CN105340008B - The compression through exploded representation of sound field - Google Patents

The compression through exploded representation of sound field Download PDF

Info

Publication number
CN105340008B
CN105340008B CN201480031162.XA CN201480031162A CN105340008B CN 105340008 B CN105340008 B CN 105340008B CN 201480031162 A CN201480031162 A CN 201480031162A CN 105340008 B CN105340008 B CN 105340008B
Authority
CN
China
Prior art keywords
vector
coefficient
described
non
spherical harmonic
Prior art date
Application number
CN201480031162.XA
Other languages
Chinese (zh)
Other versions
CN105340008A (en
Inventor
尼尔斯·京特·彼得斯
迪潘让·森
Original Assignee
高通股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US201361828445P priority Critical
Priority to US201361828615P priority
Priority to US61/828,615 priority
Priority to US61/828,445 priority
Priority to US61/829,174 priority
Priority to US61/829,182 priority
Priority to US201361829155P priority
Priority to US201361829174P priority
Priority to US201361829182P priority
Priority to US61/829,155 priority
Priority to US61/829,791 priority
Priority to US61/829,846 priority
Priority to US201361829791P priority
Priority to US201361829846P priority
Priority to US61/886,617 priority
Priority to US61/886,605 priority
Priority to US201361886605P priority
Priority to US201361886617P priority
Priority to US61/899,034 priority
Priority to US61/889,041 priority
Priority to US201361899034P priority
Priority to US201361889041P priority
Priority to US61/925,074 priority
Priority to US61/925,126 priority
Priority to US201461925126P priority
Priority to US201461925112P priority
Priority to US201461925158P priority
Priority to US201461925074P priority
Priority to US61/925,112 priority
Priority to US61/925,158 priority
Priority to US201461933721P priority
Priority to US61/933,706 priority
Priority to US61/933,721 priority
Priority to US201461933706P priority
Priority to US62/003,515 priority
Priority to US201462003515P priority
Priority to US14/289,551 priority
Priority to US14/289,551 priority patent/US9502044B2/en
Application filed by 高通股份有限公司 filed Critical 高通股份有限公司
Priority to PCT/US2014/040025 priority patent/WO2014194090A1/en
Publication of CN105340008A publication Critical patent/CN105340008A/en
Application granted granted Critical
Publication of CN105340008B publication Critical patent/CN105340008B/en

Links

Abstract

In general, present invention description is for obtaining the technology through decomposing version of spherical harmonic coefficient.According to these technologies, the device including one or more processors be can be configured with the first non-null set of the coefficient for determining the vector for the distinct components for indicating sound field, and the vector is decomposed from the multiple spherical harmonic coefficients for describing the sound field.

Description

The compression through exploded representation of sound field

Present application advocates the equity of following United States provisional application: the 61/828,445th filed on May 29th, 2013 No. 61/829,791 United States provisional application, November 1 in 2013 filed in number United States provisional application, on May 31st, 2013 No. 61/899,034 United States provisional application filed in day, No. 61/899,041 U.S. faces filed on November 1st, 2013 When application case, No. 61/829,182 United States provisional application filed on May 30th, 2013, filed on May 30th, 2013 No. 61/829,155 United States provisional application filed in No. 61/829,174 United States provisional application, on May 30th, 2013, No. 61/933,706 United States provisional application filed on January 30th, 2014, the 61/829th filed on May 31st, 2013, No. 61/886,605 United States provisional application filed in No. 846 United States provisional applications, on October 3rd, 2013,2013 years 10 No. 61/886,617 United States provisional application filed in months 3 days, No. 61/925,158 U.S. filed on January 8th, 2014 No. 61/933,721 United States provisional application filed in Provisional Application, on January 30th, 2014, filed on January 8th, 2014 No. 61/925,112 United States provisional application filed in No. 61/925,074 United States provisional application, on January 8th, 2014, No. 61/925,126 United States provisional application filed on January 8th, 2014, the 62/003rd filed on May 27th, 2014, 61/828th, No. 615 United States provisional application filed in No. 515 United States provisional applications and on May 29th, 2013, it is above The full content of each of United States provisional application is incorporated herein by reference.

Technical field

The present invention relates to audio datas, and the compression of more particularly to audio data.

Background technique

High-order ambiophony (HOA) signal (often by multiple spherical harmonic coefficients (SHC) or other class type element representations) is sound The three dimensional representation of field.This HOA or SHC expression can be independently of to reset the multi-channel audio signal from this SHC signal reproduction The modes of local loudspeaker geometrical arrangements indicate this sound field.This SHC signal may additionally facilitate backward compatibility, because can be by this SHC signal reproduction is multi channel format that is well-known and being widely adopted (for example, 5.1 voice-grade channel formats or 7.1 audios Channel format).SHC indicates therefore can realize the more preferable expression to sound field, also adapts to backward compatibility.

Summary of the invention

In general, the technology of compression and the decompression for higher-order ambiophony audio data is described.

In an aspect, a kind of method includes obtaining one or more primary vectors of the distinct components of description sound field and retouching State one or more secondary vectors of the background component of the sound field, one or more described primary vectors and it is described one or more second Both vectors are generated at least through transformation is executed relative to multiple spherical harmonic coefficients.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to really Surely describe one or more primary vectors of the distinct components of sound field and the background component of the description sound field one or more second Vector, both one or more described primary vectors and one or more secondary vectors are at least through relative to the humorous system of multiple balls Number executes transformation and generates.

In another aspect, a kind of device include: for obtain description sound field distinct components one or more first to Measure and describe the device of one or more secondary vectors of the background component of the sound field, one or more described primary vectors and described One or more secondary vectors are generated at least through transformation is executed relative to multiple spherical harmonic coefficients;And for storing described one Or the device of multiple primary vectors.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors to obtain one or more primary vectors for describing the distinct components of sound field and description when being executed One or more secondary vectors of the background component of the sound field, one or more described primary vectors and it is described one or more second to Both amounts are generated at least through transformation is executed relative to multiple spherical harmonic coefficients.

In another aspect, a kind of method include: based on indicate sound field spherical harmonic coefficient through compressed version whether be from Instruction that Composite tone object generates and one of select multiple decompression schemes;And using in the multiple decompression scheme Selected person decompress the spherical harmonic coefficient through compressed version.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to base Whether what it is in the spherical harmonic coefficient for indicating sound field is to select multiple decompressions from the instruction that Composite tone object generates through compressed version One of contracting scheme, and using the selected person in the multiple decompression scheme decompress the spherical harmonic coefficient through compressed version This.

In another aspect, a kind of device include: for based on indicate sound field spherical harmonic coefficient through compressed version whether It is the device for one of selecting multiple decompression schemes from the instruction that Composite tone object generates;And for using described more Selected person in a decompression scheme decompresses the device through compressed version of the spherical harmonic coefficient.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors of integrated decoding apparatus based on the spherical harmonic coefficient for indicating sound field through compressed version when being executed Whether this is one of to select multiple decompression schemes from the instruction that Composite tone object generates, and use the multiple solution Selected person in compression scheme decompress the spherical harmonic coefficient through compressed version.

In another aspect, a kind of method includes obtaining to indicate whether the spherical harmonic coefficient of sound field is to produce from Composite tone object Raw instruction.

In another aspect, a kind of device includes being configured to obtain to indicate whether the spherical harmonic coefficient of sound field is from synthesized voice One or more processors for the instruction that frequency object generates.

In another aspect, a kind of device includes: the device for storing the spherical harmonic coefficient for indicating sound field;And for obtaining The spherical harmonic coefficient whether be from Composite tone object generate instruction device.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors to obtain when being executed indicates whether the spherical harmonic coefficient of sound field is to generate from Composite tone object Instruction.

In another aspect, a kind of method include: one or more components of quantization means sound field one or more first to Amount;And compensation be attributed to the quantization of one or more primary vectors and in one or more the identical components for also illustrating that the sound field One or more secondary vectors in the error that introduces.

In another aspect, a kind of device includes one or more processors, one or more described processor amounts of being configured to Change one or more primary vectors for indicating one or more components of sound field, and compensates and be attributed to one or more primary vectors The error for quantifying and being introduced in one or more secondary vectors for one or more the identical components for also illustrating that the sound field.

In another aspect, a kind of device include: one or more components for quantization means sound field one or more The device of one vector;And for compensate be attributed to the quantization of one or more primary vectors and in the phase for also illustrating that the sound field With the device of the error introduced in one or more secondary vectors of one or more components.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more primary vectors of one or more components of one or more processor quantization means sound fields when being executed, and Compensation be attributed to the quantization of one or more primary vectors and the one of one or more the identical components for also illustrating that the sound field Or the error introduced in multiple secondary vectors.

In another aspect, a kind of method includes being executed based on targeted bit rates relative to multiple spherical harmonic coefficients or its decomposition Degree reduction with generate through reduction spherical harmonic coefficient or its through reduction decompose, wherein the multiple spherical harmonic coefficient indicate sound field.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to base Degree reduction is executed to generate through reduction spherical harmonic coefficient or its warp relative to multiple spherical harmonic coefficients or its decomposition in targeted bit rates Reduction is decomposed, wherein the multiple spherical harmonic coefficient indicates sound field.

In another aspect, a kind of device includes: for storing multiple spherical harmonic coefficients or its device decomposed;And it is used for base In targeted bit rates relative to the multiple spherical harmonic coefficient or its decompose execute Degree reduction with generate through reduction spherical harmonic coefficient or Its device decomposed through reduction, wherein the multiple spherical harmonic coefficient indicates sound field.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors to be based on targeted bit rates relative to multiple spherical harmonic coefficients or its decomposition execution rank when being executed Number reduction with generate through reduction spherical harmonic coefficient or its through reduction decompose, wherein the multiple spherical harmonic coefficient indicate sound field.

In another aspect, a kind of method includes the first non-zero for obtaining the coefficient of vector for the distinct components for indicating sound field Set, the vector are decomposed from multiple spherical harmonic coefficients of description sound field.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to obtain It must indicate the first non-null set of the coefficient of the vector of the distinct components of sound field, the vector is from the multiple of the description sound field Spherical harmonic coefficient decomposes.

In another aspect, a kind of device includes: for the coefficient for obtaining the vector for the distinct components for indicating sound field The device of one non-null set, the vector are decomposed from the multiple spherical harmonic coefficients for describing the sound field;And for packing coefficient The device of the first non-null set.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors to determine the first non-null set for indicating the coefficient of the vector of different component of sound field when being executed, The vector is decomposed from the multiple spherical harmonic coefficients for describing the sound field.

In another aspect, a kind of method includes obtaining to decompose from the spherical harmonic coefficient recombinated with background spherical harmonic coefficient from bit stream At least one of one or more vectors, wherein the spherical harmonic coefficient describes sound field, and wherein the background spherical harmonic coefficient is retouched State one or more background components of identical sound field.

In another aspect, a kind of device includes one or more processors, one or more described processors be configured to from Bit stream determines at least one of one or more vectors decomposed from the spherical harmonic coefficient recombinated with background spherical harmonic coefficient, wherein described Spherical harmonic coefficient describes sound field, and wherein the background spherical harmonic coefficient describes one or more background components of identical sound field.

In another aspect, a kind of device includes for obtaining from bit stream from the spherical harmonic coefficient recombinated with background spherical harmonic coefficient The device of at least one of one or more vectors decomposed, wherein the spherical harmonic coefficient describes sound field, and the wherein background Spherical harmonic coefficient describes one or more background components of identical sound field.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors to obtain from bit stream from the spherical harmonic coefficient decomposition recombinated with background spherical harmonic coefficient when being executed At least one of one or more vectors, wherein the spherical harmonic coefficient describes sound field, and the wherein background spherical harmonic coefficient description One or more background components of identical sound field.

In another aspect, a kind of method include based on the directionality determined for one or more of audio object from One or more associated spherical harmonic coefficients (SHC) of the audio object identify one or more different audio objects.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to base In the directionality determined for one or more of audio object from the humorous system of one or more balls associated with the audio object Number (SHC) identifies one or more different audio objects.

In another aspect, a kind of device includes: the device for storing one or more spherical harmonic coefficients (SHC);And it is used for Based on the directionality determined for one or more of audio object from it is associated with the audio object it is described one or more Spherical harmonic coefficient (SHC) identifies the device of one or more different audio objects.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction cause when being executed one or more processors based on the directionality determined for one or more of audio object from institute It states one or more associated spherical harmonic coefficients (SHC) of audio object and identifies one or more different audio objects.

In another aspect, a kind of method includes: to execute the synthesis based on vector relative to multiple spherical harmonic coefficients to generate Indicate the multiple spherical harmonic coefficient of one or more audio objects and corresponding direction information through exploded representation, wherein the ball is humorous Coefficient is associated with an order and describes sound field;Different and background directional information is determined from the directional information;Reduction with it is described The order of the associated directional information of background audio object is to generate transformed background directional information;Using compensation to increase through becoming The value of directional information is changed to save the gross energy of the sound field.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to: phase Executing the synthesis based on vector for multiple spherical harmonic coefficients indicates one or more audio objects and corresponding direction information to generate The multiple spherical harmonic coefficient through exploded representation, wherein the spherical harmonic coefficient is associated with an order and description sound field;From described Directional information determines different and background directional information;Reduce the order of associated with background audio object directional information with Generate transformed background directional information;Using compensation to increase the value of transformed directional information to save the total energy of the sound field Amount.

In another aspect, a kind of device include: for relative to multiple spherical harmonic coefficients execute the synthesis based on vector with The device through exploded representation for indicating the multiple spherical harmonic coefficient of one or more audio objects and corresponding direction information is generated, Described in spherical harmonic coefficient it is associated with an order and description sound field;For determining that different and background direction is believed from the directional information The device of breath;For reducing the order of directional information associated with the background audio object to generate transformed background direction The device of information;And for application compensation to increase the value of transformed directional information to save the dress of the gross energy of the sound field It sets.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors when being executed: execute the synthesis based on vector relative to multiple spherical harmonic coefficients is indicated with generating One or more audio objects and the multiple spherical harmonic coefficient of corresponding direction information through exploded representation, wherein the spherical harmonic coefficient And description sound field associated with an order;Different and background directional information is determined from the directional information;Reduction and the background The order of the associated directional information of audio object is to generate transformed background directional information;And application compensation is transformed to increase The value of directional information is to save the gross energy of the sound field.

On the other hand, a kind of method includes decomposing at least partially through first relative to more than first a spherical harmonic coefficients And second of a spherical harmonic coefficient more than second decomposes the interpolated spherical harmonic coefficient through decomposing for executing interpolation and obtaining a time slice.

In another aspect, a kind of device includes one or more processors, one or more described processors be configured to It is decomposed at least partially through first relative to more than first a spherical harmonic coefficients and the second decomposition of a spherical harmonic coefficient more than second executes Interpolation and the interpolated spherical harmonic coefficient through decomposing for obtaining a time slice.

In another aspect, a kind of device includes: for storing more than first a spherical harmonic coefficients and more than second a spherical harmonic coefficients Device;And at least partially through relative to more than described first a spherical harmonic coefficients first decompose and a ball is humorous more than second The second of coefficient decomposes the device through the interpolated spherical harmonic coefficient decomposed for executing interpolation and obtaining a time slice.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors to decompose at least partially through first relative to more than first a spherical harmonic coefficients when being executed And second of a spherical harmonic coefficient more than second decomposes the interpolated spherical harmonic coefficient through decomposing for executing interpolation and obtaining a time slice.

In another aspect, a kind of method includes the bit stream through compressed version for obtaining the spatial component including sound field, institute Stating spatial component is generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to obtain The bit stream through compressed version of the spatial component including sound field is obtained, the spatial component is by holding relative to multiple spherical harmonic coefficients Row the synthesis based on vector and generate.

In another aspect, a kind of device includes: the position through compressed version for obtaining the spatial component including sound field The device of stream, the spatial component are generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients;And it is used for Store the device of the bit stream.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described It includes the bit stream through compressed version of the spatial component of sound field, the sky that instruction causes one or more processors to obtain when being executed Between component be to be generated and executing synthesis based on vector relative to multiple spherical harmonic coefficients.

In another aspect, a kind of method includes the bit stream through compressed version for generating the spatial component including sound field, institute Stating spatial component is generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to produce Raw includes the bit stream through compressed version of the spatial component of sound field, and the spatial component is by holding relative to multiple spherical harmonic coefficients Row the synthesis based on vector and generate.

In another aspect, a kind of device includes: the position through compressed version for generating the spatial component including sound field The device of stream, the spatial component are generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients;And it is used for Store the device of the bit stream.

In another aspect, a kind of non-transitory computer-readable storage media has instruction, and described instruction is when being executed Causing one or more processors to generate includes the bit stream through compressed version of the spatial component of sound field, and the spatial component is to pass through The synthesis based on vector is executed relative to multiple spherical harmonic coefficients and is generated.

In another aspect, a kind of method includes based on multiple spatial components through in compression space component through compressed version This is relative to the multiple order identification Huffman (Huffman) codebook through the rest in compression space component to institute Use when unziping it through compressed version of spatial component is stated, the spatial component is by holding relative to multiple spherical harmonic coefficients Row the synthesis based on vector and generate.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to base In multiple spatial components through in compression space component through compressed version relative to the multiple through in compression space component Order identification Huffman (Huffman) codebook of the rest is to the spatial component when unziping it through compressed version It uses, the spatial component is generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

In another aspect, a kind of device includes: for based on multiple spatial components through in compression space component Huffman (Huffman) codebook is identified relative to the multiple order through the rest in compression space component through compressed version With in the device used when being unziped it through compressed version to the spatial component, the spatial component be by relative to Multiple humorous systems of ball execute the synthesis based on vector and generate;And for storing the multiple device through compression space component.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors based on multiple spatial components through in compression space component through compressed version when being executed Relative to the multiple order identification huffman codebook through the rest in compression space component to the spatial component Use when being unziped it through compressed version, the spatial component is by executing relative to multiple spherical harmonic coefficients based on vector Synthesis and generate.

In another aspect, a kind of method includes based on the spatial component in multiple spatial components relative to the multiple sky Between the rest in component order identification huffman codebook with the use when being compressed to the spatial component, the space Component is generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to base Spatial component in multiple spatial components identifies Huffman code relative to the order of the rest in the multiple spatial component For book with the use when compressing to the spatial component, the spatial component is by executing base relative to multiple spherical harmonic coefficients In vector synthesis and generate.

In another aspect, a kind of device includes: the device for storing huffman codebook;And for being based on multiple spaces Spatial component in component relative to the order of the rest in the multiple spatial component identify the huffman codebook with The device used when compressing to the spatial component, the spatial component are by executing base relative to multiple spherical harmonic coefficients In vector synthesis and generate.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors based on the spatial component in multiple spatial components relative to the multiple space when being executed The order identification huffman codebook of the rest in component is with the use when compressing to the spatial component, the space point Amount is generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

In another aspect, a kind of method includes determining the quantization that will be used when the spatial component to sound field compresses Step-length, the spatial component are generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

In another aspect, a kind of device includes one or more processors, one or more described processors are configured to really It is scheduled on the quantization step that will be used when compressing to the spatial component of sound field, the spatial component is by relative to multiple balls Humorous coefficient executes the synthesis based on vector and generates.

In another aspect, a kind of device includes: that will use for determining when the spatial component to sound field compresses Quantization step device, the spatial component is produced and executing synthesis based on vector relative to multiple spherical harmonic coefficients It is raw;And the device for storing the quantization step.

In another aspect, a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described Instruction causes one or more processors to determine the quantization used step when the spatial component to sound field is compressed when being executed Long, the spatial component is generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

The details of the one or more aspects of the technology is illustrated in the accompanying drawings and the description below.Other spies of these technologies Sign, target and advantage will be apparent from the description and schema and from claims.

Detailed description of the invention

Fig. 1 and 2 is the figure for the humorous basis function of ball that explanation has various orders and sub- order.

Fig. 3 is the figure of the system for the various aspects that technology described in the present invention can be performed in explanation.

Fig. 4 is described in more detail shown in the example of the Fig. 3 for the various aspects that technology described in the present invention can be performed The block diagram of one example of audio coding apparatus.

Fig. 5 is the block diagram that the audio decoding apparatus of Fig. 3 is described in more detail.

Fig. 6 is to illustrate that the content analysis unit of audio coding apparatus executes the various aspects of technology described in the present invention Example operation flow chart.

Fig. 7 is to illustrate that audio coding apparatus executes the various aspects of the synthetic technology described in the present invention based on vector Example operation flow chart.

Fig. 8 is the example operation for illustrating audio decoding apparatus in the various aspects for executing technology described in the present invention Flow chart.

Fig. 9 A to 9L is the block diagram that the various aspects of the audio coding apparatus of example of Fig. 4 are described in more detail.

Figure 10 A to 10O (ii) is the bit stream being described in more detail may specify through compression space component or side channel information A part figure.

Figure 11 A to 11G is the frame that the various units of audio decoding apparatus shown in the example of Fig. 5 are described in more detail Figure.

Figure 12 is the figure of the example audio ecosystem for the various aspects that technology described in the present invention can be performed in explanation.

Figure 13 is the figure that an example of the audio ecosystem of Figure 12 is described in more detail.

Figure 14 is the figure that an example of the audio ecosystem of Figure 12 is described in more detail.

Figure 15 A and 15B are the figures that other examples of the audio ecosystem of Figure 12 are described in more detail.

Figure 16 is the figure of the example audio code device for the various aspects that technology described in the present invention can be performed in explanation.

Figure 17 is the figure that an example of audio coding apparatus of Figure 16 is described in more detail.

Figure 18 is the figure of the example audio decoding device for the various aspects that technology described in the present invention can be performed in explanation.

Figure 19 is the figure that an example of audio decoding apparatus of Figure 18 is described in more detail.

Figure 20 A to 20G is that the example audio for the various aspects that technology described in the present invention can be performed in explanation obtains dress The figure set.

Figure 21 A to 21E is that the example audio for the various aspects that technology described in the present invention can be performed in explanation resets dress The figure set.

Figure 22 A to 22H be illustrate according to the present invention described in one or more technologies example audio playback environment Figure.

Figure 23 be illustrate according to the present invention described in one or more technologies example service condition figure, wherein user The 3D sound field of competitive sports can be experienced while wearing head-telephone.

Figure 24 be illustrate at which can according to the present invention described in one or more technical notes 3D sound field stadium Figure.

Figure 25 be illustrate according to the present invention described in one or more technologies for based on local audio scape (landscape) flow chart of the technology of 3D sound field is reproduced.

Figure 26 be illustrate according to the present invention described in one or more technologies example game studios figure.

Figure 27 be illustrate according to the present invention described in one or more technologies multiple game systems comprising reproduction engine The figure of system.

Figure 28 be illustrate according to the present invention described in one or more technologies can by head-telephone simulate loudspeaking The figure of device configuration.

Figure 29 be illustrate according to the present invention described in one or more technologies can be used to obtain and/or edit 3D sound field Multiple mobile devices figure.

Figure 30 be explanation with can according to the present invention described in the 3D sound field that is pocessed of one or more technologies it is associated Video frame figure.

Figure 31 A to 31M is to illustrate that the various aspects of technology described in displaying according to the present invention execute the synthesis of sound field Or the figure of the curve graph of the various analog results of record sort.

Figure 32 is the s-matrix for illustrating the technology described according to the present invention and decomposing from higher-order ambiophony coefficient Singular value curve graph figure.

Figure 33 A and 33B are to illustrate to show the technology described according to the present invention to the prospect component of description sound field The figure of the response curve figure of potential impact possessed by reordering when vector is encoded.

Figure 34 and 35 is to illustrate the identification according to the present invention that be based solely on energy to different audio object and be based on side The concept map of difference between the identification of tropism.

Figure 36 A to 36G be illustrate according to the present invention described in technology various aspects spherical harmonic coefficient through decomposing At least part of version projects to execute the figure of interpolation into spatial domain.

Figure 37 illustrates the expression of the technology for obtaining space-time interpolation as described herein.

Figure 38 is the artificial US matrix (US for illustrating the sequentially SVD block for multidimensional signal according to the techniques described herein1 And US2) block diagram.

Figure 39 be illustrate according to the present invention described in technology use the smooth of singular value decomposition and spatial temporal components Come decompose higher-order ambiophony (HOA) signal subsequent frame block diagram.

Figure 40 A to 40J, which is respectively, illustrates the various aspects that technology described in the present invention can be performed with packed description two The block diagram of the example audio code device of the spherical harmonic coefficient of dimension or three-dimensional sound field.

Figure 41 A to 41D is that the various aspects of technology described in the present invention can be performed to description two dimension in respective explanation Or the block diagram of example audio decoding device that the spherical harmonic coefficient of three-dimensional sound field is decoded.

The Degree reduction unit shown in the example of respective for more details Figure 40 B to the 40J of Figure 42 A to 42C Block diagram.

The figure of V compression unit shown in Figure 40 I is described in more detail in Figure 43.

Figure 44 be illustrate according to the present invention described in technology various aspects by audio coding apparatus execute with mend Repay the figure of the example operation of quantization error.

If Figure 45 A and 45B be illustrate according to the present invention described in technology various aspects the cadre from two frames Divide the figure of interpolation subframe.

Figure 46 A to 46E be illustrate according to the present invention described in technology interpolation multiple spherical harmonic coefficients through decomposing The figure of the cross section of the projection of one or more vectors of version.

Figure 47 is the frame that the extraction unit of audio decoding apparatus shown in the example of Figure 41 A to 41D is described in more detail Figure.

Figure 48 is the audio reproduction unit that audio decoding apparatus shown in the example of Figure 41 A to 41D is described in more detail Block diagram.

Figure 49 A to 49E (ii) is to illustrate that the respective audio of the various aspects of implementable technology described in the present invention is translated The figure of code system.

Figure 50 A and 50B be respective explanation according to the present invention described in technology potentially reduce the order of background content One of two distinct methods block diagram.

Figure 51 is to illustrate that the various aspects of implementable technology described in the present invention are compiled with the audio for compressing spherical harmonic coefficient The block diagram of the example of the distinct components compressed path of code device.

Figure 52 is to illustrate the various aspects of implementable technology described in the present invention to rebuild or approximatively to rebuild ball humorous The block diagram of another example of the audio decoding apparatus of coefficient (SHC).

Figure 53 is another reality of the audio coding apparatus for the various aspects that technology described in the present invention can be performed in explanation The block diagram of example.

Figure 54 is the block diagram that the example implementation of audio coding apparatus shown in the example of Figure 53 is described in more detail.

Figure 55 A and 55B are to illustrate to execute the various aspects of technology described in the present invention to rotate the example of sound field Figure.

Figure 56 is the figure for illustrating the example sound field captured according to the first referential, and first referential is then according to this hair Technology rotation described in bright is to express sound field according to the second referential.

Figure 57 A to 57E be respectively illustrate according to the present invention described in technology formed bit stream figure.

Figure 58 is the rotation that audio coding apparatus shown in the example of explanatory diagram 53 implements technology described in the present invention Turn the flow chart of the example operation of aspect.

Figure 59 is the change that audio coding apparatus shown in the example of explanatory diagram 53 executes technology described in the present invention Change the flow chart of the example operation of aspect.

Specific embodiment

The evolution of surround sound has made many output formats can be used for entertaining now.These consumption-orientation surround sound formats Example is largely " channel " formula, this is because its feed-in for being impliedly assigned to loudspeaker with certain geometric coordinates.These Comprising popular 5.1 formats (it includes following six channels: left front (FL), it is right before (FR), center or central front, it is left back or Around the left side, it is right after or around the right and low-frequency effect (LFE)), 7.1 formats of development, including, for example, 7.1.4 format and 22.2 The various formats of the height speakers such as format (for example, for being used together with ultra high-definition television standard).Non-consumption type format can Any number of loudspeaker (at symmetrical and asymmetric geometrical arrangements) are included, are usually referred to as " around array ".Such array One example includes that 32 at the coordinate on the turning for being located in and cutting icosahedron (truncated icosohedron) amplify Device.

Input to the following mpeg encoder is optionally one of three possible formats: (i) is traditional based on channel Audio (as discussed above), be intended to by preassigned position loudspeaker play;(ii) object-based Audio is related to having the associated metadata containing its position coordinates (and other information) for single audio object Discrete pulse-code modulation (PCM) data;And the audio of (iii) based on scene, be related to using the humorous basis function of ball coefficient (also referred to as Sound field is indicated for " spherical harmonic coefficient " or SHC, " high-order ambiophony " or HOA and " HOA coefficient ").This following mpeg encoder It is described in greater detail in International Organization for standardization/International Electrotechnical Commission (ISO)/(IEC) JTC1/SC29/WG11/N13411 In the document of entitled " it is required that proposal (Call for Proposals for 3D Audio) for 3D audio ", the text It offers and is issued in January, 2013 in Geneva, Switzerland, and can behttp://mpeg.chiariglione.org/sites/ default/files/files/standards/parts/docs/w13411.zipIt obtains.

There are the various formats of " surround sound " based on channel in the market.Their range is (for example) from 5.1 family's shadows Department's system (its make living room enjoy stereo aspect obtained maximum success) is to NHK (Japan Broadcasting Association or Japan Broadcast Company) 22.2 systems developed.Creator of content (for example, Hollywood studios) will wish that the track for generating film is primary, It is directed to each speaker configurations without requiring efforts and is mixed (remix) again to it.Recently, standards development organizations (Standards Developing Organizations) is considering always such as under type: coding and subsequent decoding (are suitble to In and the loudspeaker geometrical arrangements (and number) and acoustic condition that are unaware of at replay position (being related to reconstructor)) be provided in standard Change in bit stream.

To provide such flexibility to creator of content, layering elements combination can be used to indicate sound field.The layering is wanted Element set can refer to wherein element and be ordered such that the basis set of lower-order element provides the complete representation of modelling sound field One group of element.When the set expansion is with comprising higher order element, expression becomes in more detail, to increase resolution ratio.

An example for being layered elements combination is one group of spherical harmonic coefficient (SHC).Following formula demonstration is using SHC to sound field Description or expression:

This expression formula is shown can be byUniquely indicate time t sound field at any pointPlace Pressure Pi.Herein,C is the velocity of sound (about 343m/s),For reference point (or observation point), jn() is tool There is the spherical surface Bessel function (spherical Bessel function) of order n, andFor with order n and son The humorous basis function of the ball of order m.

It can be appreciated that item in square brackets be signal frequency domain representation (that is,), it can be by various It is next approximate that T/F converts (for example, discrete Fourier transform (DFT), discrete cosine transform (DCT) or wavelet transformation).Point Other examples of layer set include the set and other multiresolution basis function coefficient sets of wavelet conversion coefficient.

Fig. 1 is the figure for illustrating the humorous basis function of ball from zeroth order (n=0) to fourth order (n=4).As can be seen, for every Single order, there are the extensions of the sub- rank of m, for the purpose of ease of explanation, show the sub- rank in the example of fig. 1 but are not known and infuse It releases.

Fig. 2 is another figure for illustrating the humorous basis function of ball from zeroth order (n=0) to quadravalence (n=4).In Fig. 2, three It ties up and shows the humorous basis function of ball in coordinate space, which show both ranks and sub- rank.

It can be configured by various microphone arrays and physically obtain (for example, record)Alternatively,It being exported based on channel or object-based description from sound field.SHC indicates the audio based on scene, wherein SHC can be input to audio coder to obtain encoded SHC, and the encoded SHC can facilitate more effectively transmitting or storage.It lifts For example, it can be used and be related to (1+4)2The quadravalence of a (25, and be therefore fourth order) coefficient indicates.

As noted above, microphone can be used to record export SHC from microphone.How can be exported from microphone array The various examples of SHC are described in Poletti, M. " based on the surrounding sound system (Three-Dimensional that ball is humorous Surround Sound Systems Based on Spherical Harmonics) " (sense of hearing engineering science association proceedings (J.Audio Eng.Soc.), volume 53, o. 11th, in November, 2005, the 1004-1025 pages) in.

To illustrate to consider following equation how from object-based description these SHC of export.It can will correspond to individual sounds The coefficient of the sound field of frequency objectExpression are as follows:

Wherein i isFor spherical surface Hunk (Hankel) function (second species) with rank n, andFor the position of object.Know the source energy g (ω) become with frequency (for example, analyzing skill using T/F Art, such as Fast Fourier Transform (FFT) is executed to PCM stream) allow us to be converted to every PCM object and its positionIn addition, can show (since above formula is linear and Orthogonal Decomposition): every an objectCoefficient tool is added Property.By this method, many PCM objects can be byCoefficient (for example, summation of the coefficient vector as individual objects) carrys out table Show.Substantially, these coefficients contain the information (pressure become with 3D coordinate) about sound field, and above situation expression is being observed Point It is neighbouring slave individual objects to the transformation of the expression of whole sound field.Hereafter based on object and based on the sound of SHC Remaining each figure of the described in the text up and down of frequency decoding.

Fig. 3 is the figure for illustrating can be performed the system 10 of the various aspects of technology described in the present invention.Such as the example of Fig. 3 Shown in, system 10 includes creator of content 12 and content consumer 14.Although in creator of content 12 and content consumer 14 Described in the text up and down, but can indicate encoded in the SHC (it is alternatively referred to as HOA coefficient) of wherein sound field or any other stratum Implement the technology in any context to form the bit stream for indicating audio data.In addition, creator of content 12 can indicate energy Enough implement any type of computing device of technology described in the present invention, includes mobile phone (or cellular phone), plate meter Calculation machine, smart phone or desktop computer (several examples are provided).Equally, content consumer 14 can indicate that the present invention can be implemented Described in technology any type of computing device, comprising hand-held set (or cellular phone), tablet computer, intelligence electricity Words, set-top box or desktop computer (several examples are provided).

Creator of content 12 can indicate movie studio or can produce multichannel audio content for by such as content consumption Other entities of the content consumers such as person 14 consumption.In some instances, creator of content 12 can indicate to wish to compress HOA system The individual user of number 11.Usually, this creator of content generates audio content together with video content.The expression of content consumer 14 possesses Or the individual with the access right to audio playback systems, the audio playback systems can refer to reproduce SHC using as more sound Any type of audio playback systems that audio content is reset.In the example of fig. 3, content consumer 14 includes audio playback System 16.

Creator of content 12 includes audio editing system 18.Creator of content 12 obtain various formats (comprising directly as HOA coefficient) document recording 7 and audio object 9, audio editing system 18 can be used to edit it for creator of content 12. Creator of content can reproduce the HOA coefficient 11 from audio object 9 during editing process, to listen to reproduced loudspeaker It feeds to attempt to identify the various aspects for the sound field for needing further to edit.Creator of content 12 can then edit HOA coefficient 11 (may export for mode described above different persons in the audio object 9 of source HOA coefficient via manipulation and indirectly Editor).Audio editing system 18 can be used to generate HOA coefficient 11 in creator of content 12.The expression of audio editing system 18 can Editing audio data and export any system of this audio data as one or more source spherical harmonic coefficients.

When editing process is completed, creator of content 12 can generate bit stream 21 based on HOA coefficient 11.That is, content Founder 12 includes audio coding apparatus 20, indicates to be configured to various aspects coding in accordance with the techniques described in this disclosure Or compress HOA coefficient 11 in other ways to generate the device of bit stream 21.Audio coding apparatus 20 can produce bit stream 21 for (making For an example) emit across transmission channel, the transmission channel can be wired or wireless channel, data storage device etc..Position Stream 21 can indicate the encoded version of HOA coefficient 11, and may include that (it is logical that it can be described as side for primary bitstream and another side bit stream Road information).

Although more fully hereinafter being described, audio coding apparatus 20 be can be configured based on the synthesis based on vector Or the composite coding HOA coefficient 11 based on direction.With the determining synthetic method executed based on vector or based on the conjunction in direction At method, audio coding apparatus 20 can be at least partially based on HOA coefficient 11 and determine HOA coefficient 11 via the natural record of sound field (for example, document recording 7) still artificially (that is, synthetically) is produced from (as an example) audio object 9 such as PCM object It is raw.When HOA coefficient 11 is generated from audio object 9, the synthetic method based on direction can be used to encode HOA for audio coding apparatus 20 Coefficient 11.When HOA coefficient 11 is captured using for example intrinsic microphone (eigenmike) lively, audio coding apparatus 20 can HOA coefficient 11 is encoded based on the synthetic method based on vector.Above-mentioned difference expression can wherein be disposed based on vector or based on direction Synthetic method an example.There may be other situations: wherein any one of described synthetic method or both can be used for Naturally the mixing (mixing content) of record, artificially generated interior perhaps two kinds contents.Further it is also possible to use two kinds of sides simultaneously Method is used for the single time frame coding to HOA coefficient.

Assume for purposes of illustration: audio coding apparatus 20 determines that HOA coefficient 11 is captured or in other ways lively It indicates document recording (for example, document recording 7), audio coding apparatus 20 can be configured to use and be related to Linear Invertible Transforms (LIT) synthetic method based on vector of application encodes HOA coefficient 11.One example of Linear Invertible Transforms is referred to as " unusual Value is decomposed " (or " SVD ").In this example, SVD can be applied to HOA coefficient 11 to determine HOA coefficient by audio coding apparatus 20 11 through decompose version.The version of being decomposed that audio coding apparatus 20 can then analyze HOA coefficient 11 can promote progress to identify The various parameters to reorder through decomposing version of HOA coefficient 11.Audio coding apparatus 20 can be then based on the parameter identified The version of being decomposed of HOA coefficient 11 is reordered, wherein as described in further detail below, the case where giving following scenario described Under, this, which reorders, can improve decoding efficiency: transformation can reorder the frame of HOA coefficient across HOA coefficient, and (wherein a frame usually wraps M sample of coefficient containing HOA 11 and in some instances, 1024) M is set to.By HOA coefficient 11 through decomposing version weight After sequence, audio coding apparatus 20 can the selection of HOA coefficient 11 indicate that the prospect of sound field is (different, account for or, in other words It is advantage or outstanding) component through decompose version.Audio coding apparatus 20 can be by the warp of the expression prospect component of HOA coefficient 11 It decomposes version and is appointed as audio object and associated directional information.

Audio coding apparatus 20 can also execute HOA coefficient 11 Analysis of The Acoustic Fields at least partly to identify HOA coefficient Those of one or more background (or, in other words, environment) components of sound field HOA coefficient is indicated in 11.It is assumed that in some examples In, background component can only include the subset of any given sample of HOA coefficient 11 (for example, for example corresponding to zero and first order ball Those of face basis function sample rather than correspond to those of second order or higher order spherical substrate function sample, then audio coding Device 20 can execute energy compensating for background component.In other words, when executing Degree reduction, audio coding apparatus 20 can expand Increase the remaining background HOA coefficient in (for example, addition energy subtracts energy in) HOA coefficient 11 to compensate due to executing rank The change of total energy caused by number reduction.

Audio coding apparatus 20 next can be relative to each of HOA coefficient 11 for indicating background component and foreground voice Each of frequency object execute a form of psychologic acoustics coding (for example, MPEG is surround, MPEG-AAC, MPEG-USAC or The psychologic acoustics of other known form encodes).Audio coding apparatus 20 can execute a form of relative to developing direction information Interpolation, and then Degree reduction is executed to generate the developing direction information through Degree reduction about interpolated developing direction information. In some instances, audio coding apparatus 20 can further relate to the developing direction information through Degree reduction and execute quantization, thus Output is through decoding developing direction information.In some cases, this quantization may include scalar/entropy quantization.Audio coding apparatus 20 can Bit stream 21 is subsequently formed comprising encoded background component, encoded prospect audio object and quantified directional information.Audio Code device 20 can be transmitted then or bit stream 21 is output to content consumer 14 in other ways.

Although being shown as being transmitted directly to content consumer 14 in Fig. 3, bit stream 21 can be output to by creator of content 12 The intermediate device being located between creator of content 12 and content consumer 14.This intermediate device can store bit stream 21 for later It is delivered to the content consumer 14 that can request this bit stream.The intermediate device may include file server, network server, desk-top Computer, laptop computer, tablet computer, mobile phone, smart phone, or bit stream 21 can be stored for audio decoder Any other device that device is retrieved later.This intermediate device can reside in can (and may be in conjunction with the corresponding view of transmitting by bit stream 21 Frequency data bit stream) crossfire to request bit stream 21 subscriber's (such as content consumer 14) content delivery network in.

Alternatively, creator of content 12 can by 21 storage to storage media of bit stream, such as compact disk, digital video disk, HD video CD or other storage media, wherein most of can be read by computer and therefore can be described as computer-readable deposit Store up media or non-transitory computer-readable storage media.In this context, transmission channel can be referred to arrive so as to transmitting storage Those of content of these media channel (and may include retail shop and other delivery mechanisms based on shop).In any situation Under, technology of the invention therefore the example that in this regard should not necessarily be limited by Fig. 3.

As Fig. 3 example in further show, content consumer 14 include audio playback systems 16.Audio playback systems 16 Any audio playback systems of multichannel audb data can be indicated can to reset.Audio playback systems 16 may include it is several it is different again Existing device 22.Reconstructor 22 can respectively provide various forms of reproductions, wherein various forms of reproductions may include executing to be based on One in one or more of the various modes of the amplitude translation (VBAP) of vector, and/or the various modes of execution sound field synthesis Or more persons.As used herein, " A and/or B " means " A or B ", or both " A and B ".

Audio playback systems 16 can further include audio decoding apparatus 24.Audio decoding apparatus 24 can indicate to be configured to The device of the HOA coefficient 11' from bit stream 21 is decoded, wherein HOA coefficient 11' can be similar to HOA coefficient 11 but be attributed to damage Operation (for example, quantization) and/or different via the transmitting of transmission channel.That is, audio decoding apparatus 24 can decode bit stream The developing direction information specified in 21 carries out de-quantization, while also for the prospect audio object specified in bit stream 21 and indicating to carry on the back The encoded HOA coefficient of scape component executes psychologic acoustics decoding.Audio decoding apparatus 24 can be further relative to decoded prospect Directional information executes interpolation, and then determines expression prospect based on decoded prospect audio object and interpolated developing direction information The HOA coefficient of component.Audio decoding apparatus 24 then based on the identified HOA coefficient for indicating prospect component and can indicate background The decoded HOA coefficient of component determines HOA coefficient 11'.

Audio playback systems 16 then can decode bit stream 21 to obtain HOA coefficient 11' and reproduce HOA coefficient 11' to export Loudspeaker feeding 25.Loudspeaker feeding 25 can drive one or more loudspeakers, and (its purpose for ease of illustration is in the example of fig. 3 It is not shown).

It to select appropriate reconstructor, or is in some cases the appropriate reconstructor of generation, audio playback systems 16 can get Indicate the loudspeaker information 13 of the number of loudspeaker and/or the space geometry arrangement of loudspeaker.In some cases, audio playback System 16 reference microphone can be used to obtain loudspeaker information 13 and drive institute in a manner of dynamically determining loudspeaker information 13 State loudspeaker.In other cases or combination is dynamically determined loudspeaker information 13, and audio playback systems 16 can prompt user and sound Frequency playback system 16 interfaces with and input loudspeaker information 16.

Audio playback systems 16 then can select one of audio reproduction device 22 based on loudspeaker information 13.In some feelings Under condition, a certain threshold value class for the measurement that audio playback systems 16 can specify not in loudspeaker information 13 in audio reproduction device 22 When interior like property measurement (according to loudspeaker geometrical arrangements), audio playback systems 16 can generate audio reproduction based on loudspeaker information 13 The one in device 22.Audio playback systems 16 can generate audio reproduction device 22 based on loudspeaker information 13 in some cases In the one, without first attempt to selection audio reproduction device 22 in existing one.

Institute in the example of Fig. 3 of the various aspects of technology described in the present invention can be performed in Fig. 4 for more details The block diagram of one example of the audio coding apparatus 20 of displaying.Audio coding apparatus 20 include content analysis unit 26, based on to The synthetic method unit 27 of amount and synthetic method unit 28 based on direction.

Content analysis unit 26 indicates the content for being configured to analysis HOA coefficient 11 to identify that HOA coefficient 11 is indicated from reality The unit for the content that the content that condition record generates still is generated from audio object.Content analysis unit 26 can determine HOA coefficient 11 It is to generate from the record of practical sound field or generated from artificial audio object.Content analysis unit 26 can carry out this in various ways It determines.For example, content analysis unit 26 can be to (N+1)2- 1 channel is decoded and predicts last remaining channel (its It is represented by vector).Scalar can be applied to (N+l) by content analysis unit 262It at least some of channel-l and will be produced Raw value is added with the last remaining channel of determination.In addition, in this example, content analysis unit 26 can determine predicted channel Accuracy.In this example, if the accuracy in predicted channel is relatively high (for example, accuracy is more than specific threshold), that HOA coefficient 11 may be to generate from Composite tone object.In contrast, if the accuracy in predicted channel is relatively low (for example, accuracy is lower than specific threshold), then HOA coefficient 11 is more likely to indicate recorded sound field.For example, real herein In example, if more than 100 decibels of the signal-to-noise ratio (SNR) (dbs) in predicted channel, HOA coefficient 11 may relatively be indicated from conjunction The sound field generated at audio object.In contrast, the SNR of the sound field recorded using intrinsic microphone can be 5db to 20db.Cause And by from the SNR ratio between the sound field that the HOA coefficient 11 that practical directly record generates and generates from Composite tone object indicates Obvious boundary may be present in rate.

More precisely, when whether determine the HOA coefficient 11 for indicating sound field is to generate from Composite tone object, content point Analysis unit 26 can get the frame of HOA coefficient, indicate that (that is, N=4) can multiply 1024 for size 25 for quadravalence.Obtaining frame formula (it can also be denoted herein as frame formula SHC matrix 11 to HOA coefficient, and subsequent frame formula SHC matrix is represented by frame formula SHC matrix 27B, 27C etc.) after.Content analysis unit 26 then can exclude the primary vector of frame formula HOA coefficient 11 to generate the frame of reduction Formula HOA coefficient.In some instances, from frame formula HOA coefficient 11 exclude this primary vector can correspond in HOA coefficient 11 with Zeroth order, zero sub- those associated HOA coefficients of the humorous basis function of rank ball.

Content analysis unit 26 then can reduce frame formula HOA coefficient from the residual vector prediction for reducing frame formula HOA coefficient First non-vanishing vector.First non-vanishing vector can be referred to reach from single order (and each of consideration interdependent temper rank of order) Quadravalence each of (and consider the interdependent temper rank of order) and value are different from zero primary vector.In some instances, reduce First non-vanishing vector of frame formula HOA coefficient refer in HOA coefficient 11 it is associated with single order, the zero sub- humorous basis function of rank ball that A little HOA coefficients.Although being described about the first non-vanishing vector, the technology can from reduce frame formula HOA coefficient it is remaining to Amount prediction reduces other vectors of frame formula HOA coefficient.For example, content analysis unit 26 is predictable reduces frame formula HOA coefficient In with single order, the humorous basis function of the first sub- rank ball or associated those the HOA coefficients of single order, the humorous basis function of negative one rank ball. As other example, content analysis unit 26 is predictable reduce in frame formula HOA coefficient with second order, the humorous substrate letter of zeroth order ball Those associated HOA coefficients of number.

To predict the first non-vanishing vector, content analysis unit 26 can be operated according to following equation:

Wherein i is from 1 to (N+1)2- 2,23, α are expressed as quadravalenceiIndicate a certain constant for being used for the i-th vector, and viRefer to I vector.After predicting the first non-vanishing vector, content analysis unit 26 can be based on the first non-vanishing vector and reality of prediction Border non-vanishing vector and obtain an error.In some instances, content analysis unit 26 is subtracted from practical first non-vanishing vector through pre- The first non-vanishing vector is surveyed to export the error.Content analysis unit 26 can by error calculation be predicted first non-vanishing vector with The summation of the absolute value of the difference between each entry in practical first non-vanishing vector.

Once obtaining the error, content analysis unit 26 can be based on the energy of practical first non-vanishing vector and the error Measure calculating ratio.Content analysis unit 26 can be squared by each entry to the first non-vanishing vector and will be through squared item Phase Calais determines this energy to mesh each other.Content analysis unit 26 can then compare this ratio and a threshold value.When the ratio does not surpass Out when threshold value, content analysis unit 26 can determine that frame formula HOA coefficient 11 is to generate from record, and refer in bit stream and show HOA coefficient 11 correspondence indicate to be generated from record through decoding.When the ratio is more than threshold value, content analysis unit 26 can determine frame formula HOA coefficient 11 be generated from Composite tone object, and refer in bit stream show the correspondence of frame formula HOA coefficient 11 through decoding indicate be from Composite tone object generates.

It may include single for each frame that frame formula HOA coefficient 11, which is from the instruction that record or Composite tone object generate, Position.The single position can indicate different coding for each frame, thus effectively in the difference so as to being encoded to corresponding frame Bifurcation triggering between mode.In some cases, when frame formula HOA coefficient 11 is generated from record, content analysis unit 26 will HOA coefficient 11 is transmitted to the synthesis unit 27 based on vector.In some cases, when frame formula HOA coefficient 11 is from Composite tone When object generates, HOA coefficient 11 is transmitted to the synthesis unit 28 based on direction by content analysis unit 26.Synthesis based on direction Unit 28 can indicate the list that the bit stream 21 based on direction is generated based on the synthesis in direction for being configured to execute HOA coefficient 11 Member.

In other words, the technology is based on decoding HOA coefficient using front end classifier.The classifier can following work Make:

It is opened with frame formula SH matrix (for example 4 rank, frame sign 1024 are also known as frame formula HOA coefficient or HOA coefficient) Beginning-wherein obtains the matrix of size 25 × 1024.

Exclude the 1st vector (0 rank SH) -- therefore there are the matrixes that size is 24 × 1024.

The first non-vanishing vector (vector of 1 × 1024 size) in prediction matrix -- from its complement vector (size in matrix 23 vectors for 1 × 1024).

Predict as follows: (wherein summation I is carried out predicted vector=summation-i [α-i x vector-I] via 23 indexes, i =1...23)

It is then checked for error: the practical predicted vector=error of vector-.

If vector/error energy ratio is big (that is, error is small), basic sound field (at the frame) be it is sparse/ Synthesis.Otherwise, basic sound field is the sound field that records (for example, using microphone array).

Depending on recording to synthesis decision, (it can be related to bandwidth reduction) is encoded/decoded in different ways.Decision For 1 bit decisions sent for each frame via bit stream.

As shown in the example of fig. 4, the synthesis unit 27 based on vector may include Linear Invertible Transforms (LIT) unit 30, Parameter calculation unit 32, the unit 34 that reorders, foreground selection unit 36, energy compensating unit 38, psychologic acoustics tone decoder Unit 40, bitstream producing unit 42, Analysis of The Acoustic Fields unit 44, coefficient reduce unit 46, background (BG) selecting unit 48, space- Temporal interpolation unit 50 and quantifying unit 52.

Linear Invertible Transforms (LIT) unit 30 receives the HOA coefficient 11 in HOA channel form, and each channel indicates and ball (it is represented by HOA [k], and wherein k can be indicated for the given order of face basis function, the block of the associated coefficient of sub- order or frame The present frame or block of sample).The matrix of HOA coefficient 11 can have dimension D: M × (N+1)2

That is, LIT unit 30 can indicate the unit for being configured to execute the analysis for the form for being referred to as singular value decomposition.Although Described relative to SVD, but can relative to provide array linearly any similar transformation of incoherent energy-intensive output or It decomposes and executes technology described in the present invention.Also, " non-zero " set is intended to refer to the reference of " set " in the present invention (unless specifically state otherwise), and be not intended to refer to the classical mathematics definition of the set comprising so-called " null set ".

Alternative transforms may include the principal component analysis of often referred to as " PCA ".PCA is referred to can by one group using orthogonal transformation The observed result of energy correlated variables is converted into the mathematical routine of one group of linear uncorrelated variables referred to as principal component.Linear not phase Closing variable indicates the variable for each other and not having linear statistical relationship (or dependence).These principal components can be described as having each other There is the statistic correlation of small degree.Under any circumstance, the number of so-called principal component is less than or equal to the number of original variable. In some instances, definition transformation as follows: the first factor has maximum possible variance (or, in other words, as far as possible Mostly explain the changeability in data), and each subsequent component has possible highest variance (under following constraint: this company again Continuous component is orthogonal to (can be restated as uncorrelated in) aforementioned component).A form can be performed in PCA, according to HOA coefficient 11 It can produce the compression of HOA coefficient 11.Depending on context, PCA can be referred to by several different names, such as discrete karhunen-Lip river Transformation, Hotelling transform, appropriate Orthogonal Decomposition (POD) and eigen value decomposition (EVD) are tieed up, is named just a few.Be conducive to compress audio The characteristic of such operation of the elementary object of data is " energy compression " and " decorrelation " of multichannel audb data.

Under any circumstance, LIT unit 30 executes singular value decomposition (it can be described as " SVD " again) with by HOA coefficient 11 It is transformed to two or more set of transformed HOA coefficient.These " set " of transformed HOA coefficient may include transformed The vector of HOA coefficient.In the example in figure 4, LIT unit 30 can execute SVD relative to HOA coefficient 11 to generate so-called V square Battle array, s-matrix and U matrix.In linear algebra, SVD can be indicated to multiply z real number or complex matrix X by y by following form that (wherein X can table Show multichannel audb data, such as HOA coefficient 11) Factorization:

X=USV*

U can indicate that y multiplies y real number or complex unit matrix, and wherein the y column of U are commonly known as a left side for multichannel audb data Singular vector.S can indicate that the y with nonnegative real number number multiplies z rectangle diagonal matrix on the diagonal, wherein the diagonal line of S Value is commonly known as the singular value of multichannel audb data.V* (it can indicate the conjugate transposition of V) can indicate that z multiplies z real number or multiple Number unit matrix, wherein the z column of V* are commonly known as the right singular vector of multichannel audb data.

Although being described as being applied to the multichannel audb data comprising HOA coefficient 11 in the present invention, the technology can be answered For any type of multichannel audb data.By this method, audio coding apparatus 20 can be relative at least the one of expression sound field Partial multichannel audb data executes singular value decomposition, to generate the U square for the left singular vector for indicating multichannel audb data The V matrix of the right singular vector of battle array, the s-matrix of the singular value of expression multichannel audb data and expression multichannel audb data, And multichannel audb data is expressed as at least part of function of one or more of U matrix, s-matrix and V matrix.

In some instances, the V* matrix in above-mentioned SVD mathematic(al) representation be expressed as the conjugate transposition of V matrix with Reflection SVD can be applied to include plural matrix.When being applied to only includes the matrix of real number, the complex conjugate of V matrix (or is changed Sentence is talked about, V* matrix) transposition of V matrix can be considered as.Hereinafter it is easy the purpose of explanation, it is assumed that HOA coefficient 11 includes real Number, the result is that via SVD rather than V* Output matrix V matrix.In addition, although being expressed as V matrix in the present invention, to V matrix Refer to and be interpreted as being related to the transposition of V matrix in appropriate circumstances.Although it is assumed that V matrix, the technology can be with class It is applied to the HOA coefficient 11 with complex coefficient like mode, wherein the output of SVD is V* matrix.Therefore, in this regard, described Technology, which should not necessarily be limited by, only to be provided using SVD to generate V matrix, but may include that SVD is applied to the HOA system with complex number components Number 11 is to generate V* matrix.

Under any circumstance, LIT unit 30 can (wherein this solid be mixed relative to high-order ambiophony (HOA) audio data Ring the block or sample that audio data includes the multichannel audb data of HOA coefficient 11 or any other form) each piece (it can Referred to as frame) execute block-by-block form SVD.As noted above, parameter M can be used to indicate the length of audio frame (with sample number Meter).For example, when audio frame includes 1024 audio samples, M is equal to 1024.Although this representative value relative to M is subject to Description, but technology of the invention should not necessarily be limited by this representative value of M.LIT unit 30 can be therefore relative to multiplying (N+1) with M2It is a The block of the HOA coefficient 11 of HOA coefficient executes block SVD district by district, and wherein N indicates the order of HOA audio data again.LIT unit 30 V matrix, s-matrix and U matrix can be generated via this SVD is executed, wherein each of matrix can indicate as described above Corresponding V, S and U matrix.By this method, Linear Invertible Transforms unit 30 can execute SVD relative to HOA coefficient 11 and have to export Dimension D:M × (N+1)2US [k] vector 33 (it can indicate the group form a version of S vector sum U vector), and have dimension D:(N+ 1)2×(N+1)2V [k] vector 35.Respective vectors element in US [k] matrix is also referred to as XPS(k), and in V [k] matrix Respective vectors be also referred to as v (k).

U, the analysis of S and V matrix can be shown, these matrixes carry or indicate the space of basic sound field represented by X above And time response.Each of N number of vector in U (length is M sample) can be indicated according to the time (for by M sample The time cycle of expression) through normalized separating audio signals, it is orthogonal and (it can also quilt with any spatial character Referred to as directional information) decoupling.Representation space shape and positionThe spatial character of width can be changed to by V matrix Individual i-th vector vs(i)(k) (each has length (N+1)2) indicate.Vector in U matrix and V matrix through normalization and So that its root mean square energy is equal to unit.The energy of audio signal in U is thus indicated by the diagonal entry in S.By U and S It is multiplied to be formed US [k] (with respective vectors element XPS(k)), thus indicate have real energy audio signal.SVD is decomposed The ability for decoupling audio time signal (in U), its energy (in S) with its spatial character (in V) can be supported to be retouched in the present invention The various aspects for the technology stated.In addition, this model of basic HOA [k] coefficient X is synthesized with the vector multiplication of V [k] by US [k] Provide the term term " synthetic method based on vector " used through this document.

It is executed although depicted as directly with respect to HOA coefficient 11, but Linear Invertible Transforms can be applied to by LIT unit 30 The export item of HOA coefficient 11.For example, LIT unit 30 can be answered relative to the power spectral density matrix derived from the HOA coefficient 11 Use SVD.Power spectral density matrix is represented by PSD and obtains via the matrix multiplication of the transposition of hoaFrame to hoaFrame , it is summarized in pseudo-code as follows.HoaFrame notation refers to the frame of HOA coefficient 11.

LIT unit 30 can get S [k] after SVD (svd) is applied to PSD2Matrix (S_squared) and V [k] Matrix.S[k]2Matrix can indicate square of S [k] matrix, therefore square root calculation can be applied to S [k] by LIT unit 302Matrix To obtain S [k] matrix.In some cases, LIT unit 30 can execute quantization relative to V [k] matrix to obtain quantified V [k] Matrix (it is represented by V [k] ' matrix).LIT unit 30 can by first by S [k] Matrix Multiplication with quantified V [k] ' matrix with It obtains SV [k] ' matrix and obtains U [k] matrix.LIT unit 30 next the puppet (pinv) reciprocal of available SV [k] ' matrix and Then HOA coefficient 11 is reciprocal to obtain U [k] matrix multiplied by the puppet of SV [k] ' matrix.Afore-mentioned can be indicated by following pseudo-code:

PSD=hoaFrame'*hoaFrame;

[V, S_squared]=svd (PSD, ' econ');

S=sqrt (S_squared);

U=hoaFrame*pinv (S*V');

Pass through the power spectral density (PSD) relative to HOA coefficient rather than coefficient itself and execute SVD, LIT unit 30 can located Manage device circulation and memory space one or more in terms of possibly reduce the computational complexity for executing SVD, while realizing identical Source audio code efficiency, as SVD is directly applied to HOA coefficient.That is, above-mentioned PSD type SVD can potentially demand compared with It is few to calculate, because SVD is completed in F*F matrix (the wherein number that F is HOA coefficient).Compared with M*F matrix, wherein M is frame Length, i.e. 1024 or more samples.Via being applied to PSD rather than HOA coefficient 11, with O (M* when being applied to HOA coefficient 11 L2) compare, the complexity of SVD now can be about O (L3) (wherein O (*) indicates that calculating common in computer science and technology is complicated The big O notation of property).

Parameter calculation unit 32 indicates the unit for being configured to calculate various parameters, the parameter such as relevance parameter (R), directional characteristic parameterWith energy response (e).Each of these parameters for present frame can indicate For R [k], θ [k],R [k] and e [k].Parameter calculation unit 32 can relative to US [k] vector 33 execute energy spectrometer and/ Or correlation (or so-called crosscorrelation) is to identify these parameters.Parameter calculation unit 32 may further determine that these ginsengs of previous frame Number, wherein previous frame parameter can be expressed as R [k-1], θ [k- based on the previous frame with US [k-1] vector sum V [k-1] vector 1]、R [k-1] and e [k-1].Parameter current 37 and preceding parameters 39 can be output to and be reordered by parameter calculation unit 32 Unit 34.

That is, parameter calculation unit 32 can be relative to each of L the first US [k] vectors 33 for corresponding to first time Energy spectrometer is executed with each of the 2nd US [k-1] vector 33 for corresponding to for the second time, to calculate the first audio frame At least part (but usually from entire) and the second audio frame a part (but usually from entire) root mean square energy and borrow This generates each of 2L energy, a L for the first audio frame the first US [k] vectors 33, and one is used for second Each of the 2nd US [k-1] vector 33 of audio frame.

In other examples, parameter calculation unit 32 can be performed each of the first US [k] vector 33 with the 2nd US Crosscorrelation between certain a part (if non-entire) of the sample set of each of [k-1] vector 33.Crosscorrelation It can be referred to the crosscorrelation as understood in signal processing technology.In other words, crosscorrelation can be related to two waveforms of measurement (herein In the case of, be defined as the discrete set of M sample) between similarity, it is stagnant with the time being applied to one of wherein Become afterwards.In some instances, to execute crosscorrelation, parameter calculation unit 32 is by more every one first US [k] vector 27 of wheel Last L sample related join is determined to the preceding L sample of each of remaining person in the 2nd US [k-1] vector 33 Number.As used herein, " by wheel " operation refers to transporting by element relative to the first element set and the progress of second element set Calculate, wherein the operation according to the sequence of set and " successively " from each of first and second element sets take out a member Element.

Parameter calculation unit 32 can also analyze V [k] and/or V [k-1] vector 35 to determine directional characteristic parameter.These sides It can provide the movement of the audio object indicated by corresponding US [k] and/or US [k-l] vector 33 and the finger of position to characterisitic parameter Show.Parameter calculation unit 32 can be by any combination of the above parameter current 37 (relative to US [k] vector 33 and/or V [k] vector 35 determine) and any combination of preceding parameters 39 (determine) that offer is arrived relative to US [k-1] vector 33 and/or V [k-1] vector 35 Reorder unit 34.

SVD, which is decomposed, does not guarantee that (it is represented by US [k-1] [p] vector and (or replaces by the pth vector in US [k-1] vector 33 It is expressed as X to generationPS (p)(k-1))) indicate audio signal/object will with by the pth vector in US [k] vector 33, (it can also table It is shown as US [k] [p] vector 33 and (or is alternatively represented as XPS (p)(k))) audio signal/object indicated (is in progress) phase at any time Together.By parameter that parameter calculation unit 32 calculates for the unit 34 that reorders audio object to reorder to indicate its nature Assessment or continuity over time.

That is, reorder unit 34 then can will each of parameter 37 from the first US [k] vector 33 according to It is secondary each of with the parameter 39 of the 2nd US [k-1] vector 33 compared with.Reordering unit 34 can be based on parameter current 37 and elder generation Preceding parameter 39 reorders to each vector in US [k] matrix 33 and V [k] matrix 35 and (as an example, uses Hungarian algorithm) with by the US of rearranged sequence [k] matrix 33'(, it can be expressed mathematically as) and rearranged sequence V [k] matrix 35'(its can mathematics be expressed as) to be output to foreground sounds (or dominant sound-PS) selecting unit 36 (" preceding Scape selecting unit 36 ") and energy compensating unit 38.

In other words, the unit 34 that reorders can indicate to be configured to reorder to the vector in US [k] matrix 33 to generate The unit of US [k] the matrix 33' of rearranged sequence.The unit 34 that reorders can reorder to US [k] matrix 33, because of US [k] vector 33 (wherein, again, each vector in US [k] vector 33, can be alternatively represented as X againPS (p)(k), it can indicate to be present in One or more different (or in other words dominant) monophonic audio objects in sound field) order can be in audio data It is different between each section.That is, in some instances, it is assumed that commonly referred to as sound of the audio coding apparatus 12 to audio data These part operations of frequency frame, then correspond to these different monophonic audio objects vector position (as derived from It is represented in US [k] matrix 33) be attributable to for SVD to be applied to the different high-lightings of frame and each audio object form of interframe and It is different in audio interframe.

Vector in US [k] matrix 33 is directly delivered to psychologic acoustics tone decoder unit 40 without by audio frame Some compression schemes (such as old edition compression scheme, in monophonic sound may be reduced by reordering to the vector in US [k] matrix 33 Frequency object across audio frame be it is continuous (by channel, in this example relative to each other by the vector in US [k] matrix 33 Location order and define) when performance it is more preferable) achievable compression degree.In addition, when not reordering, in US [k] matrix 33 The coding of vector may reduce the quality of audio data in decoding.For example, with frame by frame direct coding US [k] matrix 33 The compression realized when interior vector is compared, the AAC that can be indicated in the example of fig. 3 by psychologic acoustics tone decoder unit 40 Encoder can more effectively compress one or more vectors of the rearranged sequence in US [k] matrix 33' frame by frame.Although above in relation to AAC encoder is described by, but monophonic audio object can be specified (by logical relative to across the frame in specific order or position It is genuine) when any encoder for more preferably compressing be provided execute the technology.

By this method, the various aspects of the technology aloow audio coding apparatus 12 to reset one or more vectors Sequence (for example, the vector in US [k] matrix 33, with generate the rearranged sequence in rearranged sequence US [k] matrix 33' one or more Vector and promote whereby old edition audio coder (such as psychologic acoustics tone decoder unit 40) in US [k] matrix 33 to The compression of amount).

For example, reordering unit 34 can be based on parameter current 37 and preceding parameters 39 to the US from the first audio frame One or more vectors in [k] matrix 33 reorder, first audio frame in time in US [k-1] matrix 33 one or After the second frame that multiple secondary vectors correspond to.Although in the upper of the first audio frame in time after the second audio frame It is hereinafter described by, but the first audio frame can be in time prior to the second audio frame.Therefore, the technology should not necessarily be limited by this hair Example described in bright.

To be illustrated, the following table 1 is considered, wherein each of the p vector in US [k] matrix 33 is expressed as US [k] [p], wherein k indicates that corresponding vector is to come from kth frame or previous (k-1) frame, and p indicates the vector relative to phase unisonance (wherein US [k] matrix has (N+1) to frequency frame2A such vector) vector row.As noted above, it is assumed that N is determined as 1, So p can indicate vector one (1) to (4).

Table 1

Correlation energy Compared with the following US[k-1][1] US [k] [1], US [k] [2], US [k] [3], US [k] [4] US[k-1][2] US [k] [1], US [k] [2], US [k] [3], US [k] [4] US[k-1][3] US [k] [1], US [k] [2], US [k] [3], US [k] [4] US[k-1][4] US [k] [1], US [k] [2], US [k] [3], US [k] [4]

In table 1 above, the unit 34 that reorders will be for the energy that US [k-l] [1] calculates and for US [k] [1], US Each of [k] [2], US [k] [3], US [k] [4] calculate energy comparison, will for US [k-1] [2] calculate energy with For each of US [k] [1], US [k] [2], US [k] [3], US [k] [4] energy comparison calculated, etc..Reorder list Member 34 then can abandon one or more of the 2nd US [k-1] vector 33 of previous second audio frame (according to the time).To carry out Illustrate, consider the following table 2 for showing remaining 2nd US [k-1] vector 33:

Table 2

Associated vector Related residual vector US[k-1][1] US [k] [1], US [k] [2] US[k-1][2] US [k] [1], US [k] [2] US[k-1][3] US [k] [3], US [k] [4] US[k-1][4] US [k] [3], US [k] [4]

In the above table 2, the unit 34 that reorders can determine the energy class calculated for US [k-1] [1] based on energy comparison It is similar to the energy calculated for each of US [k] [1] and US [k] [2], the energy calculated for US [k-1] [2] is similar to For the energy that each of US [k] [1] and US [k] [2] are calculated, it is similar to and is directed to for the energy that US [k-1] [3] calculate The energy that each of US [k] [3] and US [k] [4] are calculated, and the energy calculated for US [k-1] [4] is similar to and is directed to US In some instances, further energy can be performed in the unit 34 that reorders to the energy that each of [k] [3] and US [k] [4] calculate Secondary vector each of of the analysis with each of primary vector for identifying US [k] matrix 33 with US [k-1] matrix 33 Between similarity.

In other examples, the unit 32 that reorders can be based on parameter current 37 related with crosscorrelation and preceding parameters 39 It reorders to vector.In these examples, referring back to the above table 2, the unit 34 that reorders can be based on these crosscorrelation parameters Determine the following exemplary correlation expressed in table 3:

Table 3

Associated vector It is related to the following US[k-1][1] US[k][2] US[k-1][2] US[k][1] US[k-1][3] US[k][3] US[k-1][4] US[k][4]

From the above table 3, as an example, the unit 34 that reorders determines US [k-1] [1] vector US different from position [k] [2] vector correlation, US [k-1] [2] vector US [k] [1] vector correlation different from position, US [k-1] [3] vector and position Set similar US [k] [3] vector correlation, and US [k] [4] vector correlation as US [k-1] [4] vector and position class.In other words It says, the unit 34 that reorders determines the information that reorders of description following situations: the primary vector of US [k] matrix 33 that how to reorder makes The repositioning of US [k] [2] vector is in the first row of the primary vector of US [k] matrix 33 and the repositioning of US [k] [1] vector is the In second row of one US [k] vector 33.Reordering unit 34 then can be based on this information that reorders to the first of US [k] matrix 33 Vector reorders to generate the US of rearranged sequence [k] matrix 33'.

In addition, although not showing in the example in figure 4, the unit 34 that reorders this can reorder information is provided to bit stream Generation device 42, bit stream generation device 42 can produce bit stream 21 with comprising this information that reorders, so that audio decoding apparatus (example The audio decoding apparatus 24 as shown in the example of Fig. 3 and 5) it can determine how rearranged sequence vector weight to US [k] matrix 33' It sorts to restore the vector of US [k] matrix 33.

Although being described above as executing and being related to the special parameter of energy first and subsequent crosscorrelation parameter based on analysis Two-step method, but the unit 32 that reorders can only execute this analysis only with respect to energy parameter and be reordered information with determining, only with respect to Crosscorrelation parameter executes this analysis and executes to determine the information that reorders, or relative to both energy parameter and crosscorrelation parameter The analysis (mode described above).It is not related to execute energy comparison and/or intersection in addition, the technology can be used One or both other types of process in correlation is related for determining.Therefore, in this regard, the technology should not necessarily be limited by with The example of upper elaboration.In addition, from other parameters that parameter calculation unit 32 obtains, (such as spatial position is joined derived from V vector Several or V [k] and the vector in V [k-1] correlation) it can also join with the energy and crosscorrelation obtained from US [k] and US [k-1] Number is used together (simultaneously/jointly or continuously) being correctly ordered with the vector in correct US.

As an example of the correlation for using the vector in V matrix, parameter calculation unit 34 can determine V [k] matrix 35 vector be it is relevant, such as in following table 4 specified by:

Table 4

Associated vector It is related to the following V[k-1][1] V[k][2] V[k-1][2] V[k][1] V[k-1][3] V[k][3] V[k-1][4] V[k][4]

From the above table 4, as an example, the unit 34 that reorders determines V [k-1] [1] vector V [k] different from position [2] vector correlation, V [k-1] [2] vector V [k] [1] vector correlation different from position, V [k-1] [3] vector are similar with position V [k] [3] vector correlation, and V [k] [4] vector correlation as V [k-1] [4] vector and position class.Reordering unit 34 can be defeated V [k] matrix 35' of the rearranged sequence version of the vector of V [k] matrix 35 as rearranged sequence out.

In some instances, applied to the identical vector being also applied in V matrix that reorders of the vector in US matrix. In other words, any analysis for reordering to V vector can with any analysis that US vector reorders to be used in combination. In order to illustrate wherein reordering, information is not merely joined relative to US [k] vector 35 relative to energy parameter and/or crosscorrelation The determining example of number, the unit 34 that reorders can also be similar to the mode described above in relation to V [k] vector 35 and be based on intersecting phase It closes parameter and energy parameter and executes this analysis relative to V [k] vector 35.In addition, US although [k] vector 33 and do not have any side To characteristic, but V [k] vector 35 can provide the information of the directionality about corresponding US [k] vector 33.In this sense, it reorders Unit 34 can be identified related between V [k] vector 35 and V [k-1] vector 35 based on the analysis of corresponding direction characterisitic parameter.Also It is to say, in some instances, audio object is moved or kept in sound field in a continuous manner when moving in metastable Position.V [k] matrix 35 and V [k-1] matrix 35 can be showed some known entities actual motions by the unit 34 that reorders as a result, Or those of be kept fixed in sound field vector be identified as it is relevant, thus it is related based on this directional characteristic and to US [k] vector 33 and V [k] vector 35 reorders.Under any circumstance, reordering unit 34 can be by US [k] the vector 33' and warp of rearranged sequence V [k] the vector 35' to reorder is output to foreground selection unit 36.

It is not related to execute its of one or both in energy comparison and/or crosscorrelation in addition, the technology can be used The process of its type is for determining proper order.Therefore, in this regard, the technology should not necessarily be limited by examples set forth above.

Although reordering with the vector of mirror US matrix of reordering to the vector of V matrix is described above as, certain In the case of, the mode that V vector can be different from US vector reorders, wherein can produce independent syntactic element to indicate US vector It reorders and V vector reorders.In some cases, it is assumed that V vector can be encoded without psychologic acoustics, then can not to V to Amount reorders and can only reorder to US vector.

Wherein the vector of V matrix is when intention is exchanged in space from the different embodiments that reorder of the vector of US matrix When audio object -- that is, removed original recording site (when basic sound field be nature record when) or artistic intent position (when When basic sound field is the artificial mixing of object).As an example, it is assumed that there are two audio-sources A and B, A can be to come from sound field The sound " mew mew " of the cat of " left side " part, and B can be the sound " tearful " for coming from the dog of sound field " right side " part.When the weight of V and US When sorting different, the position of two sound sources is exchanged.After exchange, A (" mew mew ") is from the right part of sound field, and B (" Wang Wang ") from the left part of sound field.

Analysis of The Acoustic Fields unit 44 can indicate to be configured to execute Analysis of The Acoustic Fields relative to HOA coefficient 11 to be possible to realize The unit of targeted bit rates 41.Analysis of The Acoustic Fields unit 44 can be determined based on this analysis and/or based on received targeted bit rates 41 Psychologic acoustics decoder executes individual total number, and (it can be environment or the total number (BG of background channelTOT) function) and before The number in scape channel (or in other words predominant pathway).The sum that psychologic acoustics decoder executes individual is represented by numHOATransportChannels.Again for targeted bit rates 41 are potentially realized, Analysis of The Acoustic Fields unit 44 be may further determine that Minimal order (the N of the sum (nFG) 45 in prospect channel, background (or in other words environment) sound fieldBGOr alternatively, MinAmbHoaOrder), the corresponding number (nBGa=of the physical channel of the minimal order of background sound field is indicated (MinAmbHoaOrder+1)2), and (it in the example in figure 4 can be common for the index (i) in the channel additional BG HOA to be sent Ground is expressed as background channel information 43).Background channel information 42 is also known as environment channel information 43.Holding comes from Each of channel of numHOATransportChannels-nBGa can be " Additional background/environment channel ", " movable base In the predominant pathway of vector ", " the movable advantage signal based on direction " or " completely inactive ".In one embodiment, this A little channel types can be to indicate the syntactic element of (for " ChannelType ") (for example, 00: Additional background channel by two positions; 01: the advantage signal based on vector;10: inactive signal;11: the signal based on direction).It can be by (MinAmbHoaOrder+1 )2The multiple of the index 00 (in the above example) of+channel type being rendered as in the bit stream for the frame provides background or ring The total nBGa of border signal.

Under any circumstance, Analysis of The Acoustic Fields unit 44 can select background (or in other words ring based on targeted bit rates 41 Border) number in channel and the number in prospect (or in other words dominant) channel, thus when targeted bit rates 41 are relatively high (for example, when targeted bit rates 41 are equal to or more than 512Kbps) more backgrounds of selection and/or prospect channel.Implement at one In example, in the header segment of bit stream, numHOATransportChannels may be set to 8, and MinAmbHoaOrder can be set It is set to 1 (it is more fully described relative to Figure 10 to 10O (ii)).Under this situation, at each frame, four channels can Be exclusively used in indicate sound field background or environment division, and other 4 channels can frame by frame change on channel type -- for example, with Make Additional background/environment channel or prospect/dominance channel.Prospect/advantage signal may be based on vector or the letter based on direction One of number, as described above.

It in some cases, can be 01 (in the above example) by the ChannelType index in the bit stream of a frame Multiple provide the sum of the advantage signal based on vector for the frame.In the embodiment above, for each additional back The corresponding informance of scape/environment channel (for example, corresponding to ChannelType 00), possible HOA coefficient (exceeding first four) can It is shown in the channel.This information for quadravalence HOA content can be instructed in 5 to between 25 index ( When minAmbHoaOrder is set as 1, first four 1 can be sent always and are arrived to 4, therefore only need to indicate one between 5 to 25 Person).

Thus 5 syntactic elements (for quadravalence content) can be used to send this information, is represented by “CodedAmbCoeffIdx”。

In a second embodiment, all prospect/advantage signals are the signal based on vector.It, can in this second embodiment By nFG=numHOATransportChannels- [(MinAmbHoaOrder+1)2The multiple of+index 00] provide prospect/excellent The sum of gesture signal.

Background channel information 43 and HOA coefficient 11 are output to background (BG) selecting unit 46 by Analysis of The Acoustic Fields unit 44, will Background channel information 43 is output to coefficient and reduces unit 46 and bitstream producing unit 42, and nFG 45 is output to foreground selection list Member 36.

In some instances, Analysis of The Acoustic Fields unit 44 can be based on point of the vector of US [k] matrix 33 and targeted bit rates 41 Analysis selection has the variable nFG number of these components of maximum value.In other words, Analysis of The Acoustic Fields unit 44 can be by analyzing by S The vector of [k] matrix 33 successively decrease diagonal line value generation slope of a curve and determine separate two sub-spaces parameter A (it can It is similar or be substantially similar to NBG) value, wherein big singular value indicates prospect or different sound, and low singular value indicates sound field Background component.That is, overall sound field can be segmented into prospect subspace and background subspace by parameter A.

In some instances, the single order and second dervative of singular value curve can be used in Analysis of The Acoustic Fields unit 44.Analysis of The Acoustic Fields The value of parameter A can be also limited between one and five by unit 44.As another example, Analysis of The Acoustic Fields unit 44 can be by parameter A's Value is limited in one and (N+l)2Between.Alternatively, the value of parameter A can be defined as in advance such as value four by Analysis of The Acoustic Fields unit 44.? In any case, based on the value of A, Analysis of The Acoustic Fields unit 44 determines the order of the sum (nFG) 45 in prospect channel, background sound field (NBG) and the number (nBGa) in the channel additional BG HOA and index (i) to be sent.

In addition, Analysis of The Acoustic Fields unit 44 can determine to every vector the energy of the vector in V [k] matrix 35.Analysis of The Acoustic Fields list Member 44 can determine the energy of each of vector in V [k] matrix 35, and before having those of high-energy vector to be identified as Scape component.

In addition, Analysis of The Acoustic Fields unit 44 can execute HOA coefficient 11 various other analyses, including dimensional energy analysis, The auditory analysis of spatial concealment analysis, diffusion analysis or other forms.Analysis of The Acoustic Fields unit 44 can be converted via by HOA coefficient 11 Dimensional energy analysis is executed to spatial domain, and identifies the high-energy regions for indicating the durection component for the sound field that should be saved.Sound field The mode that analytical unit 44 can be similar to dimensional energy analysis executes aware space masking analysis, only Analysis of The Acoustic Fields unit 44 It can recognize the area of space by spatially close higher-energy sound mask.In some cases, Analysis of The Acoustic Fields unit 44 can Then it is based on the less prospect component of perceptual mask region recognition.Analysis of The Acoustic Fields unit 44 can be executed further to HOA coefficient 11 Diffusion analysis is to identify the propagation energy region of the background component that can indicate sound field.

Analysis of The Acoustic Fields unit 44 can also indicate to be configured to using the information based on directionality associated with audio data Determine the unit of the high-lighting for indicating the audio data of sound field, diversity or dominance.Although the determination based on energy can improve By the reproduction of the SVD sound field decomposed to identify the different audio component of sound field, but show the feelings of high level in background audio component Under condition, it can also make device that background audio component are mistakenly identified as different audio component based on the determination of energy.That is, it is different and The separation based on only energy of background audio component may be unstable, because high energy (for example, louder) background audio component may Improperly it is identified as different audio component.In order to more steadily be distinguished between the different background audio component of sound field, this The various aspects of technology described in invention can make Analysis of The Acoustic Fields unit 44 be able to carry out HOA coefficient 11 based on directionality Analysis is to separate prospect and environmental audio component with the version of being decomposed of HOA coefficient 11.

In in this respect, Analysis of The Acoustic Fields unit 44 can indicate to be configured or can operate in other ways to identify from US The background element for including in one or more of vector in vector sum V [k] matrix 35 in [k] matrix 33 it is different (or preceding Scape) element unit.According to some technologies based on SVD, the highest component of energy is (for example, US [k] matrix 33 and V [k] matrix Several vectors or vector derived from it before one or more of 35) it can be treated to be distinct components.However, US [k] matrix The highest component of energy (it is indicated by vector) of one or more of the vector in vector sum V [k] matrix 35 in 33 is not Component/signal of most directionality can be indicated in all scenario.

The one or more aspects of the implementable technology described herein of Analysis of The Acoustic Fields unit 44 are to be based on US [k] matrix 33 In vector sum V [k] matrix 35 in vector or one or more of vector derived from it vector directionality and identify Prospect/directly/advantage element.In some instances, Analysis of The Acoustic Fields unit 44 can both directionality of vector based on energy know Not or select one or more vectors as different audio component (wherein the component is also known as " object ").For example, Analysis of The Acoustic Fields unit 44 can will be in the vector (or from its derived from vector) in vector sum V [k] matrix 35 in US [k] matrix 33 One or more in display high-energy and those of both high directivity (for example, being expressed as directionality quotient) vector be identified as phase Abnormal sound frequency component.As a result, if Analysis of The Acoustic Fields unit 44 determines vector sum V [k] matrix in specific vector and US [k] matrix 33 Other vectors in one or more of vector (or from its derived from vector) in 35, which are compared, has relatively small number of directionality, So regardless of energy level associated with the specific vector, Analysis of The Acoustic Fields unit 44 can determine that the specific vector indicates By background (or environment) audio component for the sound field that HOA coefficient 11 indicates.

In some instances, Analysis of The Acoustic Fields unit 44 can identify different audio pair based on directionality by executing following operation As (as noted above, alternatively referred to as " component ").Analysis of The Acoustic Fields unit 44 can (it can be from US [k] vector by S [k] matrix 33 export or (although in the example of Fig. 4 individually diagram) are exported by LIT unit 30) in vector multiplied by (for example, using one or Multiple matrix multiplication processes) vector in V [k] matrix 35.By making V [k] matrix 35 and S [k] multiplication of vectors, Analysis of The Acoustic Fields list First 44 available VS [k] matrixes.In addition, Analysis of The Acoustic Fields unit 44 can be to the entry of each of vector in VS [k] matrix At least some of it is squared (that is, power be two exponentiation).In some cases, Analysis of The Acoustic Fields unit 44 can be to each vector It is associated with the order greater than 1 those sum through squared entry.

As an example, if each vector of VS [k] matrix includes 25 entries, Analysis of The Acoustic Fields unit 44 can It is squared relative to entry of each vector to each vector for starting at fifth entry and terminating at Article 25 mesh, It sums to through squared entry to determine directionality quotient (or directional indicator).Each summation operation can lead to it is corresponding to The directionality quotient of amount.In this example, Analysis of The Acoustic Fields unit 44 can determine the associated with the order less than or equal to 1 of every a line Those of entry (that is, first arrive fourth entry), be more generally directed to the amount of energy, and the less direction for those entries Property.That is, lower-order ambiophony associated with order zero or one corresponds to Fig. 1 and spherical surface base illustrated in fig. 2 Bottom function does not provide many in terms of the direction of pressure wave, but actually provides a certain volume (it indicates energy).

Operating described in above example can also express according to following pseudo-code.Following pseudo-code include mark, in comprising The form of comment statement in character string "/* " and the continuous example of " */" (no quotation marks).

[U, S, V]=svd (audioframe, ' ecom');

VS=V*S;

/ * next row for independently analyzing every a line, and will from fifth entry to Article 25 purpose first (as One example) row in value summation with determine a corresponding vector quotient or directionality measurement.By the entry involution before summation. It is associated with higher-order ambiophony with the associated entry of rank greater than 1 in every a line, and be therefore more likely directionality 's.*/

SumVS=sum (VS (5:end :) .^2,1);

/ * next row be for the VS matrix that will generate square summation sequence, and selection maximum value set (for example, Three in maximum value or four) */

[~, idxVS]=sort (sumVS, ' descend');

U=U (:, idxVS);

V=V (:, idxVS);

S=S (idxVS, idxVS);

In other words, according to the above pseudo-code, corresponding to big from HOA coefficient 11 is may be selected in Analysis of The Acoustic Fields unit 44 In one entry of each vector of VS [k] matrix decomposed of those of the spherical basis function of order HOA coefficient.Analysis of The Acoustic Fields Single 44 can then by these entry involutions of each vector of VS [k] matrix, will identify, calculate through square entry summation or with Other way determines directionality measurement or the quotient of each vector of VS [k] matrix.Next, Analysis of The Acoustic Fields unit 44 can based on to The respective party tropism of each of amount is measured the vector order of VS [k] matrix.Analysis of The Acoustic Fields unit 44 can be by these vectors With the descending sort that directionality is measured so that have those of highest counterparty tropism vector be at first, and have it is minimum right It is last for answering those of directionality vector.Analysis of The Acoustic Fields unit 44 then can select vector have highest opposite side tropism amount The non-zero subset of degree.

Any combination of previous analysis can be performed to determine that psychologic acoustics decoder executes individual in Analysis of The Acoustic Fields unit 44 (it can be environment or the sum (BG of background channel to sumTOT) and prospect channel number function.Analysis of The Acoustic Fields unit 44 can Order (the N of the sum (nFG) 45 in prospect channel, background sound field is determined based on any combination of previous analysisBG) and to be sent (it can be referred to collectively as background channel information to the number (nBGa) and index (i) in the additional channel BG HOA in the example in figure 4 43)。

In some instances, Analysis of The Acoustic Fields unit 44 can every M sample (it can be restated as frame by frame) execute this point Analysis.In in this respect, the value of A can be in interframe difference.The example that wherein every M sample carries out the bit stream of decision is showed in Figure 10 and arrives In 10O (ii).In other examples, Analysis of The Acoustic Fields unit 44 can every frame execute this analysis more than once, thus the two of analysis frame A or more than two parts.Therefore, the technology should not necessarily be limited by example described in the present invention in this respect.

Foreground selection unit 48 can indicate to be configured to based on background channel information (for example, background sound field (NBG) and to The number (nBGa) and index (i) in the channel additional BG HOA sent) determine the unit of background or environment HOA coefficient 47.Citing For, work as NBGEqual to for the moment, the every the same of the audio frame with the order equal to or less than one is may be selected in Foreground selection unit 48 This HOA coefficient 11.In this example, Foreground selection unit 48 can be selected then with by indexing one of (i) identification The HOA coefficient 11 of index is used as additional BG HOA coefficient, generates list in place wherein the nBGa to specify in bit stream 21 is provided Member 42 is to make audio decoding apparatus (for example, the audio decoding apparatus 24 shown in the example of Fig. 3) can be from bit stream 21 Parse BG HOA coefficient 47.Environment HOA coefficient 47 then can be output to energy compensating unit 38 by Foreground selection unit 48.Environment HOA coefficient 47 can have dimension D: M × [(NBG+1)2+nBGa]。

Foreground selection unit 36 can indicate to be configured to based on nFG 45 (its can indicate to identify these prospect vectors one or Multiple indexes) US [k] the matrix 33' of the rearranged sequence of selection and the expression sound field in V [k] the matrix 35' of rearranged sequence prospect Or the unit of those of distinct components.Foreground selection unit 36 can be by (its US for being represented by rearranged sequence of nFG signal 49 [k]1 ..., nFG 49、FG1 ..., nfG[k] 49, or(k) 49) it is output to psychologic acoustics tone decoder unit 40, wherein nFG Signal 49 can have dimension D: M × nFG, and each indicates monophonic audio object.Foreground selection unit 36 can also be by correspondence In V [k] the matrix 35'(or v of the rearranged sequence of the prospect component of sound field(1..nFG)(k) 35') it is output to space-time interpolation list Member 50, wherein those of the prospect component that corresponds in V [k] the matrix 35' of rearranged sequence can be represented as having dimension D ((N+1 )2× nFG) prospect V [k] matrix 51k(it can be mathematically represented as)。

Energy compensating unit 38 can indicate to be configured to execute energy compensating relative to environment HOA coefficient 47 to compensate attribution In the unit of the energy loss generated by each in the removal of Foreground selection unit 48 channel HOA.Energy compensating unit 38 can Relative to US [k] the matrix 33' of rearranged sequence, V [k] matrix 35', nFG signal 49 of rearranged sequence, prospect V [k] vector 51k Energy spectrometer is executed with one or more of environment HOA coefficient 47, and energy compensating is then executed to produce based on this energy spectrometer The raw environment HOA coefficient 47' through energy compensating.Energy compensating unit 38 can export the environment HOA coefficient 47' through energy compensating To psychologic acoustics tone decoder member 40.

Effectively, energy compensating unit 38 can be used for compensating the context components due to reducing the sound field that HOA coefficient 11 describes Order with the environment HOA coefficient 47 that generates Degree reduction, (it has an order less than N in some instances, according to corresponding to What it is with following order/sub- order spherical substrate function uniquely includes coefficient: [(NBG+1)2+ nBGa]) caused by sound field The total energy of background sound component may be decreased.In some instances, energy compensating unit 38 is compensated by following operation This energy loss: the [(N for being suitable for environment HOA coefficient 47 is determinedBG+1)2+ nBGa] column each of the shape in value of magnification The compensating gain of formula is equal to or to increase to the root mean square (RMS) of 47 energy of environment HOA coefficient at least closer to ground approximation HOA coefficient 11 RMS (such as via US [k] the matrix 33' of rearranged sequence, rearranged sequence V [k] matrix 35', nFG signal 49, Prospect V [k] vector 51kDetermined with the total energy spectrometer of one or more of the environment HOA coefficient 47 of Degree reduction), with Environment HOA coefficient 47 is output to psychologic acoustics tone decoder unit 40 afterwards.

In some cases, energy compensating unit 38 can recognize the V of US [k] the matrix 33' and rearranged sequence of rearranged sequence The RMS of every a line and/or column on one or more of [k] matrix 35'.Energy compensating unit 38 also can recognize that selected prospect is logical (it may include nFG signal 49 and prospect V [k] vector 51 in one or more of roadkAnd the environment HOA coefficient of Degree reduction 47) RMS of every a line and/or column.It is described in US [k] the matrix 33' of rearranged sequence and V [k] the matrix 35' of rearranged sequence Every a line of one or more and/or the RMS of column can be stored as being expressed as RMSFULLVector, and nFG signal 49, prospect V [k] to Amount 51kIt can be stored as being expressed as with every a line of one or more of the environment HOA coefficient 47 of Degree reduction and/or the RMS of column RMSREDUCEDVector.Energy compensating unit 38 then can calculate value of magnification vector Z: Z=RMS according to following equationFULL/ RMSREDUCED.This value of magnification vector Z or its each section then can be applied to nFG signal 49, prospect V by energy compensating unit 38 [k] vector 51kOne or more of with the environment HOA coefficient 47 of Degree reduction.In some cases, value of magnification vector Z foundation Following equation HOABG-RED'=HOABG-REDZTIt is applied to the environment HOA coefficient 47 of only Degree reduction, wherein HOABG-REDIndicate order The environment HOA coefficient 47, HOA of reductionBG-RED'Indicate the HOA coefficient 47' of the Degree reduction through energy compensating, and ZTIndicate Z-direction amount Transposition.

In some instances, believe for V [k] matrix 35', nFG of US [k] the matrix 33' of the rearranged sequence of determination, rearranged sequence Number 49, prospect V [k] vector 51kIt is every with the corresponding line of one or more of the environment HOA coefficient 47 of Degree reduction and/or column One RMS, energy compensating unit 38 will can be applied to the column with reference to spherical harmonic coefficient (SHC) reconstructor first.By energy compensating list 38 application of member allows to determine the RMS in the domain SHC with US [k] the matrix 33' of the rearranged sequence of determination, rearranged with reference to SHC reconstructor V [k] matrix 35', nFG signal 49, prospect V [k] vector 51 of sequencekWith it is one or more in the environment HOA coefficient 47 of Degree reduction The energy of the described overall sound field of every a line and/or column for the frame that the row and/or column of person indicates, it is as follows to retouch in more detail It states.

Space-time interpolation unit 50 can indicate prospect V [k] vector 51 for being configured to receive kth framekAnd former frame Prospect V [k-1] vector 51 of (therefore being k-1 notation)k-1And execute space-time interpolation with generate interpolated prospect V [k] to The unit of amount.Space-time interpolation unit 50 can be by nFG signal 49 and prospect V [k] vector 51kIt reconfigures to restore through weight The prospect HOA coefficient of sequence.Space-time interpolation unit 50 can be then by the prospect HOA coefficient of rearranged sequence divided by interpolated V [k] vector to generate interpolated nFG signal 49'.Also exportable prospect V [k] vector 51 of space-time interpolation unit 50k For generating those of interpolated prospect V [k] vector prospect V [k] vector 51kSo that the audio such as audio decoding apparatus 24 Decoding apparatus can produce interpolated prospect V [k] vector and restore prospect V [k] vector 51 wherebyk.By prospect V [k] vector 51k's To generate those of interpolated prospect V [k] vector prospect V [k] vector 51kIt is expressed as remaining prospect V [k] vector 53.For Ensure can to compile at encoder and decoder using identical V [k] and V [k-1] (create interpolated vector V [k]) These quantified/dequantized version is used at code device and decoder.

In this regard, when space-time interpolation unit 50 can be indicated from some other parts of the first audio frame and second Between on the first audio frame of rear or preceding audio frame interpolation first part unit.In some instances, the part can It is expressed as subframe, wherein the interpolation such as executed relative to subframe is more fully described below with respect to Figure 45 to 46E.In other examples In, space-time interpolation unit 50 can be relative to the sample of a certain last number of former frame and a certain first number of subsequent frame Purpose sample operations are such as described in more detail relative to Figure 37 to 39.Space-time interpolation unit 50 can subtract when executing this interpolation Specified prospect V [k] vector 51 is needed in small bit stream 21kSample number, because of prospect V [k] vector 51kOnly those use In prospect V [k] vector 51 for generating interpolated V [k] vectorkExpression prospect V [k] vector 51kSubset.That is, in order to latent Keep the compression of HOA coefficient 11 more effective (by reducing prospect V [k] vector 51 specified in bit stream 21 on groundkNumber), this The various aspects of technology described in invention can provide the interpolation of one or more parts of the first audio frame, wherein the part Each of can indicate HOA coefficient 11 through decompose version.

Space-time interpolation can lead to several benefits.Firstly, due to execute SVD or other LIT block property district by district, NFG signal 49 can not be continuous from frame to frame.In other words, produced under conditions of LIT unit 30 applies SVD frame by frame Transformed HOA coefficient in specific discontinuity may be present, such as the unordered property institute of such as US [k] matrix 33 and V [k] matrix 35 It proves.By executing this interpolation, can have in interpolation and potentially reduce due to frame boundaries are (or in other words HOA coefficient 11 divides Section is frame) and can reduce under conditions of the smooth effect of any artifact of introducing discontinuous.Prospect of the application V [k] vector 51kIt executes This interpolation and then be based on interpolated prospect V [k] vector 51kInterpolated nFG letter is generated from the rearranged sequence HOA coefficient restored Number 49' can make to be attributed to operation frame by frame and be attributed to smooth at least some effects that nFG signal 49 reorders.

In operation, space-time interpolation unit 50 can interpolation a HOA coefficient more than first be contained in first frame The first of 11 a part is decomposed (for example, prospect V [k] vector 51k) and more than second a HOA coefficients 11 for being contained in the second frame A part second decompose (for example, prospect V [k] vector 51k-1) one or more subframes of the first audio frame be used for generating The interpolated spherical harmonic coefficient through decomposing of one or more subframes.

In some instances, before the first of the right singular vector of the part of first decomposition including indicating HOA coefficient 11 Scape V [k] vector 51k.Equally, in some instances, second decompose include indicate HOA coefficient 11 the part the right side it is unusual to Second prospect V [k] vector 51 of amountk

It in other words, can be the ginseng of 3D pressure field based on the humorous 3D audio of ball for the orthogonal basement function on spherical surface Number indicates.The order N of the expression is higher, and spatial resolution is possibly higher, and the number of usually humorous (SH) coefficient of ball is bigger ((N+1) in total2A coefficient).For many applications, it may be necessary to which the bandwidth reduction of coefficient can effectively emit and store institute State coefficient.This targeted in the present invention technology can provide to be reduced using the dimension based on frame of singular value decomposition (SVD) Journey.Each frame of coefficient can be resolved into three matrix Us, S and V by SVD analysis.In some instances, the technology can be by US [k] Some durection component disposition as basic sound field in vector in matrix.However, when disposing by this method, these vectors (in U S [k] matrix) is even if be discontinuous in interframe -- it indicates same different audio component.It is translated when by converting audio frequency When component described in code device feed-in, these discontinuities can lead to significant artifact.

Technology described in the present invention can solve this discontinuity.That is, the technology can be based on following observed result: V Matrix can the interpreted orthogonal intersection space axis in the humorous domain of ball.U [k] matrix can indicate humorous (HOA) data of ball according to those substrate letters Several projections, wherein discontinuity is attributable to orthogonal spatial axes (V [k]), the spatial axes line per frame change and therefore from As discontinuous.This is different from the similar decomposition such as Fourier transform, and wherein basis function is in some instances in interframe It will be constant.In these terms, SVD is regarded as matching and pursues algorithm.Technology described in the present invention can make interpolation list Member 50 can maintain the continuity between basis function (V [k]) in interframe and carrying out interpolation therebetween.

As noted above, interpolation can be executed relative to sample.When subframe includes the single set of sample, this situation exists Generalization is able in above description.In two situations via sample and via the interpolation of subframe, interpolative operation can be in following The form of equation:

In this above-mentioned equation, interpolation can be executed from single V- vector v (k-1) relative to single V- vector v (k), The V- vector from contiguous frames k and k-1 can be indicated in one embodiment.In above-mentioned equation, l indicates that execution interpolation is targeted Resolution ratio, wherein l can indicate integral sample and (wherein T is the length of sample, executes interpolation in the length by l=1 ..., T And the interpolated vector for needing to export in the lengthAnd the length also indicate this process output generate these to The l of amount).Alternatively, l can indicate the subframe being composed of a plurality of samples.When (for example) dividing the frame into four subframes, l can be wrapped Include the value 1,2,3 and 4 for each of the subframe.It can be via bit stream using the value of l as being referred to as The field of " CodedSpatialInterpolationTime " is issued with signal, so that interpolative operation can be repeated in a decoder. W (l) may include the value of interpolation weight.When it is interior be inserted as linear when, w (l) can be used as the function of l and linearly and monotonously 0 With 1 between change.In other cases, w (l) can be used as the function of l with non-linear but dull mode (such as the four of raised cosine / mono- circulation) change between 0 and 1.Function w (l) can be indexed between several different functions possibilities and by function w (l) it is issued as the field for being referred to as " SpatialInterpolationMethod " with signal in bit stream, so that can be by solving Code thinks highly of again identical interpolative operation.When w (l) be close to 0 value when, outputIt can largely be weighted or by v (k-1) It influences.And when w (l) be close to 1 value when, ensure to exportIt is largely weighted or is influenced by v (k-1).

Coefficient reduces unit 46 and can indicate to be configured to based on background channel information 43 relative to remaining prospect V [k] vector 53 execution coefficients are reduced so that reduced prospect V [k] vector 55 to be output to the unit of quantifying unit 52.Reduced prospect V [k] vector 55 can have dimension D:[(N+1)2-(NBG+1)2-nBGa]×nFG。

Coefficient reduces unit 46 and middle expression can be configured to reduce the coefficient of remaining prospect V [k] vector 53 in this respect The unit of number.In other words, coefficient reduces unit 46 and can indicate to be configured to eliminate (to form remaining prospect V [k] vector 53 ) prospect V [k] vector has few unit to those of zero direction information coefficient.As described above, in some examples In, (it can be indicated those of single order and zeroth order the basis function coefficient that corresponds to of phase exclusive or (in other words) prospect V [k] vector For NBG) few directional information is provided, and therefore can be removed from prospect V vector (via the process that can be referred to " coefficient reduction "). In this example, it is possible to provide larger flexibility is not only from set [(NBG+1)2+ 1, (N+1)2] identify corresponding to NBGThese be It counts and identifies the additional channel HOA (it can be indicated by variable TotalOfAddAmbHOAChan).Analysis of The Acoustic Fields unit 44 can divide HOA coefficient 11 is analysed to determine BGTOT, can not only identify (NBG+1)2And (it can unite identification TotalOfAddAmbHOAChan Referred to as background channel information 43).Coefficient, which reduces unit 46 and then can remove from remaining prospect V [k] vector 53, corresponds to (NBG+1)2 With those of TotalOfAddAmbHOAChan coefficient to generate size as ((N+1)2-(BGTOT) x nFG smaller dimension V [k] Matrix 55 is also known as prospect V [k] vector 55 of reduction.

Quantifying unit 52 can indicate to be configured to execute prospect V [k] vector 55 of any type of quantization to compress reduction To generate through decoding prospect V [k] vector 57 to which these are output to bitstream producing unit 42 through decoding prospect V [k] vector 57 Unit.In operation, quantifying unit 52 can indicate the spatial component for being configured to compression sound field (that is, in this example, reducing One or more of prospect V [k] vector 55) unit.For purposes of example, it is assumed that prospect V [k] vector 55 of reduction wraps Containing two row vectors, since coefficient is reduced, every a line has less than 25 elements (it implies that the quadravalence HOA of sound field is indicated).Although It is described relative to two row vectors, but any number vector may include in prospect V [k] amount 55 of reduction, be at most (n+ 1)2A, wherein n indicates the order that the HOA of sound field is indicated.In addition, although being hereinafter described as executing scalar and/or entropy quantization, Any type of quantization of the compression of executable prospect V [k] vector 55 for leading to reduction of quantifying unit 52.

Quantifying unit 52 can receive prospect V [k] vector 55 of reduction and execute compression scheme to generate through decoding prospect V [k] vector 57.This compression scheme usually can be related to the element for compressing vector or data any it is contemplated that compression scheme, and The example that should not necessarily be limited by following more detail.As an example, quantifying unit 52 it is executable comprising in the following one or The compression scheme of more persons: the floating point representation of each element of prospect V [k] vector 55 of reduction is transformed into the prospect V [k] of reduction The uniform quantization of the integer representation of prospect V [k] vector 55 of integer representation, the reduction of each element of vector 55, and it is remaining The classification and decoding of the quantified integer representation of prospect V [k] vector 55.

In some instances, each during one or more of this compression scheme can be controlled by dynamic state of parameters to realize Or almost realize the targeted bit rates 41 for being used for gained bit stream 21 (as an example).In prospect V [k] vector 55 of reduction Each it is orthogonal under conditions of, each of prospect V [k] vector 55 of reduction can independent interpretation.In some examples In, it is as follows described in more detail, before identical decoding mode (being defined by various subpatterns) can be used to decode each reduction Each element of scape V [k] vector 55.

Under any circumstance, as noted above, this decoding scheme can first relate to prospect V [k] vector 55 that will reduce Each of the floating point representation (it is 32 floating numbers in some instances) of each element be transformed to 16 integer representations. Quantifying unit 52 can be by each element of the given person in prospect V [k] vector 55 that will reduce multiplied by 215(in some examples In, executed and moving to right 15) and this floating-point is executed to integer transform.

Quantifying unit 52 then can execute equal relative to all elements of the given person in prospect V [k] vector 55 of reduction One quantization.Quantifying unit 52 can identify quantization step based on the value for being represented by nbits parameter.Quantifying unit 52 can be based on target Bit rate 41 dynamically determines this nbits parameter.Quantifying unit 52 can determine quantization step according to this nbits parameter.As one Quantization step (being expressed as " residual quantity " or " Δ " in the present invention) can be determined as being equal to 2 by a example, quantifying unit 5216-nbits。 In this example, if nbits is equal to six, residual quantity is equal to 210, and there are 26A quantification levels.In in this respect, for Vector element v, quantified vector element vqEqual to [v/ Δ], and -2nbits-1<vq<2nbits-1

Quantifying unit 52 can then execute the classification and remaining decoding of quantified vector element.As an example, quantization Following equation can be used to give quantified vector element v for one for unit 52qIdentify that classification corresponding to this element (passes through determination Class identifier cid):

Quantifying unit 52 then can carry out Hoffman decodeng to this classification index cid, while also identification instruction vqFor positive value Or the sign bits of negative value.

Next quantifying unit 52 can recognize the remnants in this classification.As an example, quantifying unit 52 can according to Lower equation determines this remnants:

Remaining=| vq|-2cid-1

Quantifying unit 52 then can carry out block decoding to this remnants with cid-1 position.

Following instance illustrates the simplified example of this classification and remaining decoding process.First, it is assumed that nbits, which is equal to 6, makes vq ∈ [- 31,31].Next, assuming following:

cid vq Huffman code for cid 0 0 ‘1’ 1 - 1,1 ‘01’ 2 - 3, -2,2,3 ‘000’ 3 - 7, -6, -5, -4,4,5,6,7 ‘0010’ 4 - 15, -14 ..., -8,8 ..., 14,15 ‘00110’ 5 - 31, -30 ..., -16,16 ..., 30,31 ‘00111’

And, it is assumed that it is following:

cid For remaining block code 0 It is unavailable 1 0,1 2 01,00,10,11 3 011,010,001,000,100,101,110,111 4 0111,0110..., 0000,1000 ..., 1110,1111 5 01111 ..., 00000,10000 ..., 11111

Thus, for vq=[6, -17,0,0,3], it may be determined that following:

> > cid=3,5,0,0,2

> > sign=l, 0, x, x, l

> > remaining=2,1, x, x, 1

> > is directed to 6 position=' 0010 '+' 1 '+' 10 '

> > is directed to -17 position=' 00111 '+' 0 '+' 0001 '

> > is directed to 0 position=' 0 '

> > is directed to 0 position=' 0 '

> > is directed to 3 position=' 000 '+' 1 '+' 1 '

> > total position=7+10+1+1+5=24

> > average bit=24/5=4.8

Although being not shown in previous simplified example, quantifying unit 52 can be when decoding cid for nbits's Different value selects different huffman codebooks.In some instances, quantifying unit 52 can be directed to nbits value 6 ..., and 15 provide difference Hoffman decodeng table.In addition, it is 6 that quantifying unit 52, which may include for range, ..., each of 15 different nbits values Five different huffman codebooks, in total 50 huffman codebooks.In this regard, quantifying unit 52 may include multiple and different Huffmans Codebook is to adapt to several different decodings for counting the cid in situations.

In order to illustrate quantifying unit 52 can be for each of nbits value comprising for translating vector element one to four First huffman codebook of code, is used for vector element nine the second huffman codebook for decoding to vector element five to nine The third huffman codebook decoded above.When there is following scenario described, this first three huffman codebook can be used: before reduction Prospect V [k] vector 55 to be compressed is not subsequent in time from prospect V [k] vector 55 of reduction in scape V [k] vector 55 Correspondence prospect V [k] vector forecasting and not indicate Composite tone object ((for example) initially by through pulse-code modulation (PCM) The audio object that audio object defines) spatial information.Quantifying unit 52 can include to use otherwise for each of nbits value 4th huffman codebook of the one decoding in prospect V [k] vector 55 to reduction is (when prospect V [k] vector of reduction This one in 55 is corresponding prospect V [k] vector forecasting subsequent in time from prospect V [k] vector 55 of reduction When).Quantifying unit 52 can also be for each of nbits value comprising for described in prospect V [k] vector 55 to reduction The 5th huffman codebook of one decoding (when this one in prospect V [k] vector 55 of reduction indicates Composite tone object). It can be in these difference statistics situations (that is, in this example, nonanticipating and non-synthetic situation, prediction situation and synthesis situation) Each develop various huffman codebooks.

Following table illustrates huffman table selection and to specify in bit stream so that decompression unit can select appropriate Hough The position of graceful table:

Prediction mode HT information HT table 0 0 HT5 0 1 HT{1,2,3} 1 0 HT4 1 1 HT5

In preceding table, prediction mode (" Pred mode ") indicates whether to perform prediction for current vector, and Huffman Table (" HT information ") indicates additional huffman codebook (or table) information to select one of huffman table one to five.

Following table further illustrates this huffman table selection course (it is assumed that various statistics situations or situation).

Record Synthesis Without Pred HT{1,2,3} HT5 There is Pred HT4 HT5

In preceding table, " record " column instruction vector indicates decoding situation when recorded audio object, and " synthesis " arranges Indicate decoding situation when vector indicates Composite tone object.The instruction of " no Pred " row does not execute prediction relative to vector element When decoding situation, and " having Pred " row instruction relative to vector element execute predict when decoding situation.As shown in this table, Quantifying unit 52 selects HT { 1,2,3 } when vector indicates recorded audio object and does not execute and predicts relative to vector element. Quantifying unit 52 selects HT5 when audio object indicates Composite tone object and does not execute and predicts relative to vector element.Quantization Unit 52 selects HT4 when vector indicates recorded audio object and executes and predicts relative to vector element.Quantifying unit 52 exists Audio object indicates Composite tone object and selects HT5 when executing and predict relative to vector element.

In in this respect, scalar quantization and/or huffman coding described above is can be performed to compress contracting in quantifying unit 52 Prospect V [k] vector 55 subtracted, so that output is through decoding prospect V [k] vector 57 (it is referred to alternatively as side channel information 57).This Side channel information 57 may include the syntactic element to decode to remaining prospect V [k] vector 55.Quantifying unit 52 can be similar The mode of the mode shown in the example of one of Figure 10 B and 10C exports side channel information 57.

As described above, quantifying unit 52 can produce the syntactic element of side channel information 57.For example, quantifying unit 52 It may specify which in the multiple configuration mode the expression in the header of access unit (it can include one or more of frame) select The syntactic element of configuration mode.It is specified although depicted as on the basis of every access unit, but quantifying unit 52 can be in every frame This syntactic element is specified in basic or any other periodic basis or aperiodicity basic (such as entire bit stream is primary).It is in office In the case of what, this syntactic element may include that instruction has selected which one in four configuration modes for specifying the prospect V of reduction Two positions in terms of direction of the nonzero coefficient set of [k] vector 55 to indicate this distinct components.Institute's syntax elements can indicate For " codedVVecLength ".By this method, quantifying unit 52 can be signaled in bit stream or be specified in other ways and make It is fixed through decoding prospect V [k] vector 57 to be referred in bit stream with which one in four configuration modes.Although being configured relative to four Mode is described by, but the technology should not necessarily be limited by four configuration modes, and should be limited to any number of configuration mode, comprising single A configuration mode or multiple configuration modes.Flag 63 can be also appointed as in side channel information 57 by scalar/entropy quantifying unit 53 Another syntactic element.

The psychologic acoustics tone decoder unit 40 for including in audio coding apparatus 20 can indicate psychologic acoustics audio coding Multiple examples of device, each of which is for encoding through every in energy compensating environment HOA coefficient 47' and interpolated nFG signal 49' The different audio objects of one or the channel HOA are to generate encoded environment HOA coefficient 59 and encoded nFG signal 61.Psychological sound Bitstream producing unit can be output to for encoded environment HOA coefficient 59 and encoded nFG signal 61 by learning tone decoder unit 40 42。

In some cases, this psychologic acoustics tone decoder 40 can indicate advanced audio decoding (AAC) coding unit One or more examples.Psychologic acoustics tone decoder unit 40 can be to environment HOA coefficient 47' through energy compensating and interpolated Each column or row of nFG signal 49' are encoded.Usually, psychologic acoustics tone decoder unit 40 can be for through energy compensating Environment HOA coefficient 47' and interpolated nFG signal 49' in each of remaining rank/sub- rank combination call AAC coding single The item of member.About how being found in angstrom using the more information that AAC coding unit encodes background spherical harmonic coefficient 31 The entitled of Li Kehelu moral (Eric Hellerud) et al. " encodes higher-order ambiophony (Encoding with AAC Higher Order Ambisonics with AAC) " conference paper in, in the 124th conference (on May 17th, 2008 To 20 days) on submit and can be obtained at lower place:Http:// ro.uow.edu.au/cgi/viewcontent.cgi? article =: 8025&context=engpapers.In some cases, audio coding unit 14 can be used more interpolated than for encoding NFG signal 49' low targeted bit rates to through energy compensating environment HOA coefficient 47' carry out audio coding, whereby with it is interpolated NFG signal 49', which is compared, potentially more compresses the environment HOA coefficient 47' through energy compensating.

The bitstream producing unit 42 being included in audio coding apparatus 20 is indicated data format to meet known format (it can be referred to format known to decoding apparatus) generates the unit of the bit stream 21 based on vector whereby.Bitstream producing unit 42 is one Multiplexer can be indicated in a little examples, can receive through decoding prospect V [k] vector 57, encoded environment HOA coefficient 59, warp Encode nFG signal 61 and background channel information 43.Bitstream producing unit 42 can then based on through decoding prospect V [k] vector 57, Encoded environment HOA coefficient 59, encoded nFG signal 61 and background channel information 43 generate bit stream 21.Bit stream 21 may include master It wants or status of a sovereign stream and one or more side channel bit streams.

Although not showing in the example in figure 4, audio coding apparatus 20 also may include bitstream output unit, the bit stream Output unit will use the synthesis based on direction or the composite coding based on vector based on present frame and switch from audio coding The bit stream (for example, switching between the bit stream 21 based on direction and the bit stream based on vector 21) that device 20 exports.This bit stream is defeated Unit can execute synthesizing based on direction (as detecting HOA coefficient 11 based on the instruction exported by content analysis unit 26 out The result generated from Composite tone object) or synthesis of the execution based on vector (as detecting the recorded knot of HOA coefficient Fruit) syntactic element execute this switching.Bitstream output unit may specify correct header grammer to indicate for present frame and This switching or present encoding of corresponding one in bit stream 21.

In some cases, the various aspects of the technology can also make audio coding apparatus 20 can determine HOA coefficient 11 It whether is generated from Composite tone object.These aspects of the technology can enable audio coding apparatus 20 be configured to obtain Whether the spherical harmonic coefficient for indicating sound field is the instruction generated from Composite tone object.

In these and other cases, audio coding apparatus 20 is further configured to determine whether spherical harmonic coefficient is from conjunction It is generated at audio object.

In these and other cases, audio coding apparatus 20 be configured to from storage indicate sound field spherical harmonic coefficient to At least part of frame formula spherical harmonic coefficient matrix excludes primary vector to obtain the frame formula spherical harmonic coefficient matrix of reduction.

In these and other cases, audio coding apparatus 20 be configured to from storage indicate sound field spherical harmonic coefficient to At least part of frame formula spherical harmonic coefficient matrix excludes primary vector to obtain the frame formula spherical harmonic coefficient matrix of reduction, and based on described The residual vector of the frame formula spherical harmonic coefficient matrix of reduction predicts the vector of the frame formula spherical harmonic coefficient matrix of the reduction.

In these and other cases, audio coding apparatus 20 be configured to from storage indicate sound field spherical harmonic coefficient to At least part of frame formula spherical harmonic coefficient matrix excludes primary vector to obtain through downscaled frame formula spherical harmonic coefficient matrix, and at least partly Summation prediction of the ground based on the residual vector through downscaled frame formula spherical harmonic coefficient matrix is described through downscaled frame formula spherical harmonic coefficient square The vector of battle array.

In these and other cases, audio coding apparatus 20 is configured to be based at least partially on frame formula spherical harmonic coefficient square The vector of at least part of frame formula spherical harmonic coefficient matrix of the summation prediction storage spherical harmonic coefficient of the residual vector of battle array.

In these and other cases, audio coding apparatus 20 is configured to be based at least partially on frame formula spherical harmonic coefficient square The vector of at least part of frame formula spherical harmonic coefficient matrix of the summation prediction storage spherical harmonic coefficient of the residual vector of battle array, and be based on The vector predicted calculates error.

In these and other cases, audio coding apparatus 20 is configured to be based at least partially on frame formula spherical harmonic coefficient square The vector of at least part of frame formula spherical harmonic coefficient matrix of the summation prediction storage spherical harmonic coefficient of the residual vector of battle array, and be based on The correspondence vector for the vector sum frame formula spherical harmonic coefficient matrix predicted calculates error.

In these and other cases, audio coding apparatus 20 is configured to be based at least partially on frame formula spherical harmonic coefficient square The vector of at least part of frame formula spherical harmonic coefficient matrix of the summation prediction storage spherical harmonic coefficient of the residual vector of battle array, and will accidentally Difference is calculated as the summation of institute's predicted vector with the absolute value of the difference of the corresponding vector of frame formula spherical harmonic coefficient matrix.

In these and other cases, audio coding apparatus 20 is configured to be based at least partially on frame formula spherical harmonic coefficient square The vector of at least part of frame formula spherical harmonic coefficient matrix of the summation prediction storage spherical harmonic coefficient of the residual vector of battle array, is based on institute The correspondence vector of predicted vector and frame formula spherical harmonic coefficient matrix calculates error, correspondence vector based on frame formula spherical harmonic coefficient matrix with The energy balane ratio of the error, and the ratio and threshold value indicate whether the spherical harmonic coefficient of sound field is from conjunction with determination It is generated at audio object.

In these and other cases, audio coding apparatus 20 be configured to storage spherical harmonic coefficient through compressed version The instruction is specified in bit stream 21.

In some cases, various technologies can enable audio coding apparatus 20 execute transformation relative to HOA coefficient 11.? In the case of these and other, audio coding apparatus 20 can be configured with obtain description sound field distinct components one or more first Vector sum describes one or more secondary vectors of the background component of sound field, one or more described primary vectors and it is described one or more Both secondary vectors are generated at least through transformation is executed relative to the multiple spherical harmonic coefficient 11.

In these and other cases, audio coding apparatus 20, wherein transformation includes singular value decomposition, the singular value point The S that solution generates and indicates the U matrix of the left singular vector of the multiple spherical harmonic coefficient, indicates the singular value of the multiple spherical harmonic coefficient The V matrix of the right singular vector of matrix and the multiple spherical harmonic coefficient 11 of expression.

In these and other cases, audio coding apparatus 20, wherein one or more described primary vectors include one or more A U through audio codingDIST*SDISTVector, be before audio coding by by U matrix one or more through audio coding UDISTVector multiplied by s-matrix one or more SDISTVector and generate, and wherein U matrix and s-matrix are at least through for phase Singular value decomposition is executed for the multiple spherical harmonic coefficient and is generated.

In these and other cases, audio coding apparatus 20, wherein one or more described primary vectors include one or more A U through audio codingDIST*SDISTVector sum has one or more V of the transposition of V matrixT DISTVector, the UDIST*SDISTTo Amount is before audio coding by by one or more U through audio coding of U matrixDISTVector is multiplied by the one or more of s-matrix A SDISTVector and generate, and wherein U matrix and s-matrix and V matrix are at least through relative to the multiple spherical harmonic coefficient 11 execute singular value decompositions and generate.

In these and other cases, audio coding apparatus 20, wherein one or more described primary vectors include one or more A UDIST*SDISTVector sum has one or more V of the transposition of V matrixT DISTVector, the UDIST*SDISTVector is compiled in audio By by one or more U through audio coding of U matrix before codeDISTVector multiplied by s-matrix one or more SDISTVector and produce It is raw, wherein U matrix, s-matrix and V matrix be at least through for executing singular value decomposition relative to the multiple spherical harmonic coefficient and Generate, and wherein audio coding apparatus 20 be further configured to obtain instruction will be extracted from bit stream with described in being formed one or more UDIST*SDISTOne or more V described in vector sumT DISTThe value D of the number of vectors of vector.

In these and other cases, audio coding apparatus 20, wherein one or more described primary vectors include one or more A UDIST*SDISTVector sum has one or more V of the transposition of V matrixT DISTVector, the UDIST*SDISTVector is compiled in audio By by one or more U through audio coding of U matrix before codeDISTVector multiplied by s-matrix one or more SDISTVector and produce It is raw, wherein U matrix, s-matrix and V matrix be at least through for executing singular value decomposition relative to the multiple spherical harmonic coefficient and Generate, and wherein audio coding apparatus 20 is further configured to based on audio frame acquisition value D one by one, described value D instruction to from Bit stream is extracted to form one or more described UDIST*SDISTOne or more V described in vector sumT DISTThe number of the vector of vector.

In these and other cases, audio coding apparatus 20, wherein the transformation includes principal component analysis to identify sound The distinct components of field and the background component of sound field.

The various aspects of technology described in the present invention can provide the audio coding dress for being configured to compensate for quantization error Set 20.

In some cases, audio coding apparatus 20 can be configured with the one of one or more components of quantization means sound field or Multiple primary vectors, and compensate the quantization for being attributed to one or more primary vectors and also illustrating that the mutually same of the sound field Or the error introduced in one or more secondary vectors of multiple components.

In these and other cases, audio coding apparatus is configured to quantization from least partially through relative to retouching The multiple spherical harmonic coefficients for stating sound field execute singular value decomposition and one or more vectors of the transposition of V matrix for generating.

In these and other cases, audio coding apparatus is further configured to humorous relative to multiple balls of expression sound field Coefficient executes singular value decomposition to generate the U matrix for the left singular vector for indicating the multiple spherical harmonic coefficient, indicate the multiple The s-matrix of the singular value of spherical harmonic coefficient, and indicate the V matrix of the right singular vector of the multiple spherical harmonic coefficient, and be configured to Quantify one or more vectors of the transposition from V matrix.

In these and other cases, audio coding apparatus is further configured to humorous relative to multiple balls of expression sound field Coefficient executes singular value decomposition to generate the U matrix for the left singular vector for indicating the multiple spherical harmonic coefficient, indicate the multiple The s-matrix of the singular value of spherical harmonic coefficient, and indicate the V matrix of the right singular vector of the multiple spherical harmonic coefficient, the amount of being configured to Change the transposition from V matrix one or more vectors, and be configured to compensate for be attributed to by by one or more U of U matrix to It measures multiplied by one or more S vectors of s-matrix and the quantization of one or more calculated U*S vectors and the error that introduces.

In these and other cases, audio coding apparatus is further configured to humorous relative to multiple balls of expression sound field Coefficient executes singular value decomposition to generate the U matrix for the left singular vector for indicating the multiple spherical harmonic coefficient, indicate the multiple The s-matrix of the singular value of spherical harmonic coefficient, and indicate the V matrix of the right singular vector of the multiple spherical harmonic coefficient;Determine U matrix One or more UDISTVector, each of which corresponds to the distinct components of sound field;Determine one or more S of s-matrixDISTVector, it is every One corresponds to the identical distinct components of sound field;And determine one or more V of the transposition of V matrixT DISTVector, each of which is right It should be in the identical distinct components of sound field;It is configured to quantify one or more described VT DISTVector is to generate one or more VT Q_DIST Vector;And it is configured to compensate for being attributed to by by one or more U described in U matrixDISTVector multiplied by s-matrix one or more SDISTVector is to generate one or more U through error compensationDIST*SDISTVector and one or more calculated UDIST*SDISTTo The quantization of amount and the error introduced.

In these and other cases, audio coding apparatus is configured to based on one or more described UDISTIt is vector, described One or more SDISTOne or more V described in vector sumT DISTVector determines different spherical harmonic coefficient, and relative to the VT Q_DISTVector Execute pseudoinverse with by the different spherical harmonic coefficient divided by one or more described VT Q_DISTVector and generating in turn is at least partly mended It repays and passes through VT DISTThe quantization of vector and one or more U through error compensation of error introducedC_DIST*SC_DISTVector.

In these and other cases, audio coding apparatus is further configured to humorous relative to multiple balls of expression sound field Coefficient executes singular value decomposition to generate the U matrix for the left singular vector for indicating the multiple spherical harmonic coefficient, indicate the multiple The V matrix of the right singular vector of the s-matrix and the multiple spherical harmonic coefficient of expression of the singular value of spherical harmonic coefficient, determines description sound field One or more background components U matrix one or more UBGVector sum describes the U matrix of one or more distinct components of sound field One or more UDISTVector determines one or more S of the s-matrix of one or more background components of description sound fieldBGVector With one or more S of the s-matrix of one or more distinct components described in description sound fieldDISTVector, and determine the transposition of V matrix One or more VT DISTOne or more V of vector sumT BGVector, wherein VT DISTOne or more described distinct components of vector description sound field And VT BGOne or more the described background components for describing sound field are configured to quantify one or more described VT DISTVector is to generate one Or multiple VT Q_DISTVector, and be configured to compensate for being attributed to through one or more U by described inBGVector multiplied by it is described one or more SBGVector and then multiplied by one or more described VT BGThe back that vector is formed to generate the background spherical harmonic coefficient through error compensation Quantization in scape spherical harmonic coefficient and the error introduced.

In these and other cases, audio coding apparatus is configured to based on VT DISTVector sum is by by the UDISTTo Amount is multiplied by SDISTVector and one or more U formedDIST*SDISTVector determines the error, and identified error is added to Background spherical harmonic coefficient is to generate the background spherical harmonic coefficient through error compensation.

In these and other cases, audio coding apparatus is configured to compensate for being attributed to one or more described primary vectors Quantization and the error that is introduced in one or more secondary vectors for one or more the identical components for also illustrating that sound field to generate One or more secondary vectors through error compensation, and further be configured to generate bit stream with comprising it is described one or more through error The secondary vector of compensation and described one or more quantified primary vectors.

In these and other cases, audio coding apparatus is configured to compensate for being attributed to one or more described primary vectors Quantization and the error that is introduced in one or more secondary vectors for one or more the identical components for also illustrating that sound field to generate One or more secondary vectors through error compensation, and be further configured to it is described one or more through error compensation second to Amount carries out audio coding, and generates bit stream with secondary vector comprising one or more through audio coding through error compensation and described One or more quantified primary vectors.

The various aspects of the technology can further such that audio coding apparatus 20 can generate through reduction spherical harmonic coefficient or It is decomposed.In some cases, audio coding apparatus 20 can be configured to be based on targeted bit rates relative to multiple spherical harmonic coefficients Its decompose execute Degree reduction with generate through reduction spherical harmonic coefficient or its through reduction decompose, wherein the multiple spherical harmonic coefficient table Show sound field.

In these and other cases, audio coding apparatus 20 is opposite before being further configured to execution Degree reduction In the multiple spherical harmonic coefficient execute singular value decomposition with identify description sound field distinct components one or more primary vectors and It identifies one or more secondary vectors of the background component of sound field, and is configured to relative to one or more described primary vectors, institute It states one or more secondary vectors or both one or more described primary vectors and one or more secondary vectors executes order contracting Subtract.

In these and other cases, audio coding apparatus 20 is further configured to relative to the multiple spherical harmonic coefficient Or it decomposes and executes content analysis, and is configured to humorous relative to the multiple ball based on targeted bit rates and the content analysis Coefficient or its decompose execute Degree reduction with generate through reduction spherical harmonic coefficient or its through reduction decompose.

In these and other cases, audio coding apparatus 20 is configured to relative to the multiple spherical harmonic coefficient or its point Solution executes spatial analysis.

In these and other cases, audio coding apparatus 20 is configured to relative to the multiple spherical harmonic coefficient or its point Solution executes diffusion analysis.

In these and other cases, audio coding apparatus 20 be configured to relative to the multiple spherical harmonic coefficient or its Decompose one or more processors for executing spatial analysis and diffusion analysis.

In these and other cases, audio coding apparatus 20 be further configured to comprising through reduction spherical harmonic coefficient or It, which is reduced in the bit stream decomposed, specifies through reduction spherical harmonic coefficient or its one through the corresponding spherical substrate function of reduction decomposition Or multiple orders and/or one or more sub- ranks.

In these and other cases, through reduction spherical harmonic coefficient or its through reduction decompose than the multiple spherical harmonic coefficient or its Decomposing has lesser value.

In these and other cases, audio coding apparatus 20 is configured to remove the multiple spherical harmonic coefficient or its decomposition Vector in there is specified those of order and/or sub- rank spherical harmonic coefficient or vector to generate through reduction spherical harmonic coefficient or its warp Reduction is decomposed.

In these and other cases, audio coding apparatus 20 is configured to the multiple spherical harmonic coefficient of pulverised or its decomposition Those of in vector there is specified those of order and/or sub- rank spherical harmonic coefficient or vector to generate through reduction spherical harmonic coefficient or It is decomposed through reduction.

The various aspects of the technology may also allow for audio coding apparatus 20 to be configured to indicate the distinct components of sound field.? In the case of these and other, audio coding apparatus 20 is configured to obtain is to the vector to indicate the distinct components of sound field The non-null set of several first, wherein the vector is decomposed from multiple spherical harmonic coefficients of description sound field.

In these and other cases, audio coding apparatus 20 is configured to determine the described of the coefficient of the vector First non-null set is to include all coefficients.

In these and other cases, audio coding apparatus 20 is configured to determine described the first of coefficient the non-null set It is greater than the order of basis function corresponding to one or more of the multiple spherical harmonic coefficient for corresponding in the coefficient Those of order coefficient.

In these and other cases, audio coding apparatus 20 be configured to determine the first non-null set of coefficient with comprising Corresponding in the coefficient is greater than the rank of the order of basis function corresponding to one or more of the multiple spherical harmonic coefficient Those of number coefficient, and it is right to exclude the one or more institute that corresponding in the coefficient is greater than in the multiple spherical harmonic coefficient At least one of the order of the order for the basis function answered.

In these and other cases, audio coding apparatus 20 be configured to determine the described first non-null set of coefficient with Comprising all coefficients, but correspond to the rank for being greater than basis function corresponding to one or more of the multiple spherical harmonic coefficient Except at least one of described coefficient of several orders.

In these and other cases, audio coding apparatus 20, which is further configured in side channel information, specifies institute State the described first non-null set of the coefficient of vector.

In these and other cases, audio coding apparatus 20, which is further configured in side channel information, specifies institute The the described first non-null set for stating the coefficient of vector, without the described first non-null set of the coefficient to the vector Carry out audio coding.

In these and other cases, the vector includes being divided using the synthesis based on vector from the multiple spherical harmonic coefficient The vector of solution.

In these and other cases, the synthesis based on vector includes singular value decomposition.

In these and other cases, the vector includes being decomposed using singular value decomposition from the multiple spherical harmonic coefficient V vector.

In these and other cases, audio coding apparatus 20 be further configured to select in multiple configuration modes so as to Specify a configuration mode of the non-null set of the coefficient of the vector, and based on described in the multiple configuration mode Selected person specifies the non-null set of the coefficient of the vector.

In these and other cases, the one in the multiple configuration mode indicates the non-zero of the coefficient Set includes all coefficients.

In these and other cases, the non-null set of the one instruction coefficient in the multiple configuration mode It is greater than the order of basis function corresponding to one or more of the multiple spherical harmonic coefficient comprising corresponding in the coefficient Those of order coefficient.

In these and other cases, the one in the multiple configuration mode indicates the non-zero of the coefficient Set is greater than basis function corresponding to one or more of the multiple spherical harmonic coefficient comprising corresponding in the coefficient Those of order of order coefficient, and exclude in the coefficient correspond to described one be greater than in the multiple spherical harmonic coefficient or At least one coefficient of the order of the order of the basis function corresponding to more persons.

In these and other cases, the non-null set of the one instruction coefficient in the multiple configuration mode Comprising all coefficients, but except at least one of described coefficient.

In these and other cases, audio coding apparatus 20, which is further configured in bit stream, specifies the multiple match Set the selected one in mode.

The various aspects of technology described in the present invention may also allow for audio coding apparatus 20 to be configured and with various side The distinct components of formula expression sound field.In these and other cases, audio coding apparatus 20 is configured to obtain to indicate The non-null set of the first of the coefficient of the vector of the distinct components of sound field, wherein the vector is from the humorous system of multiple balls of description sound field Number decomposes.

In these and other cases, the described first non-null set of coefficient includes all coefficients of vector.

In these and other cases, the described first non-null set of coefficient includes corresponding in the coefficient to be greater than institute State those of the order of order of basis function corresponding to one or more of multiple spherical harmonic coefficients coefficient.

In these and other cases, the described first non-null set of the coefficient includes corresponding to greatly in the coefficient Those of the order of order of basis function corresponding to one or more of the multiple spherical harmonic coefficient coefficient, and exclude institute State the institute for corresponding to the basis function corresponding to the one or more being greater than in the multiple spherical harmonic coefficient in coefficient State at least one coefficient of the order of order.

In these and other cases, the described first non-null set of coefficient includes all coefficients, but the coefficient In be identified as except at least one coefficient without enough directional informations.

In these and other cases, audio coding apparatus 20 is further configured to the described first non-null set of coefficient Close the first part for being extracted as vector.

In these and other cases, audio coding apparatus 20 is further configured to from side channel information extracted vector The first non-null set, and the described first non-null set of the coefficient based on the vector obtains the weight of the multiple spherical harmonic coefficient Group version.

In these and other cases, the vector includes being divided using the synthesis based on vector from the multiple spherical harmonic coefficient The vector of solution.

In these and other cases, the synthesis based on vector includes singular value decomposition.

In these and other cases, audio coding apparatus 20 is further configured to determine so as to according to multiple configuration moulds One of formula extracts the one in the multiple configuration mode of the non-null set of the coefficient of the vector, and is based on institute State the non-null set that the institute winner in multiple configuration modes extracts the coefficient of the vector.

In these and other cases, the one in the multiple configuration mode indicates the non-zero of the coefficient Set includes all coefficients.

In these and other cases, the non-null set of the one instruction coefficient in the multiple configuration mode It is greater than the order of basis function corresponding to one or more of the multiple spherical harmonic coefficient comprising corresponding in the coefficient Those of order coefficient.

In these and other cases, the one in the multiple configuration mode indicates the non-zero of the coefficient Set is greater than basis function corresponding to one or more of the multiple spherical harmonic coefficient comprising corresponding in the coefficient Those of order of order coefficient, and exclude in the coefficient correspond to described one be greater than in the multiple spherical harmonic coefficient or At least one coefficient of the order of the order of the basis function corresponding to more persons.

In these and other cases, the non-null set of the one instruction coefficient in the multiple configuration mode Comprising all coefficients, but except at least one of described coefficient.

In these and other cases, audio coding apparatus 20 is configured to true based on the value signaled in bit stream The one in fixed the multiple configuration mode.

In some cases, the various aspects of the technology also aloow audio coding apparatus 20 to identify one or more Different audio object (or, in other words, advantage audio object).In some cases, audio coding apparatus 20 can be configured with It is humorous from one or more balls associated with the audio object based on the directionality determined for one or more of audio object Coefficient (SHC) identifies one or more different audio objects.

In these and other cases, audio coding apparatus 20 is further configured to based on associated with audio object Spherical harmonic coefficient determines the directionality of one or more audio objects.

In these and other cases, audio coding apparatus 20 is further configured to execute relative to spherical harmonic coefficient unusual Value is decomposed to generate the U matrix for the left singular vector for indicating the multiple spherical harmonic coefficient, indicate the surprise of the multiple spherical harmonic coefficient The V matrix of the right singular vector of the s-matrix and the multiple spherical harmonic coefficient of expression of different value, and the multiple spherical harmonic coefficient is indicated For at least part of function of one or more of U matrix, s-matrix and V matrix, wherein audio coding apparatus 20 is configured to It is at least partially based on the respective party tropism that the V matrix determines one or more audio objects.

In these and other cases, audio coding apparatus 20 is further configured to one or more vectors to V matrix It reorders, so that the vector with larger directionality quotient is located in smaller directionality quotient in the V matrix of rearranged sequence On vector.

In these and other cases, audio coding apparatus 20 is further configured to determine with larger directionality quotient's Vector includes more directional informations than the vector with smaller directionality quotient.

In these and other cases, audio coding apparatus 20 is further configured to V Matrix Multiplication with s-matrix to generate VS matrix, the VS matrix include one or more vectors.

In these and other cases, audio coding apparatus 20 be further configured to select every a line of VS matrix with It is squared to each of selected entry to form corresponding squared entry and right greater than the associated entry of 14 order In every a line of VS matrix, summed to all squared entries to determine the directionality quotient of corresponding vector.

In these and other cases, audio coding apparatus 20 be configured to selection VS matrix every a line be greater than 14 The associated entry of order include selecting to start at the 18th entry of every a line of VS matrix and in every a line of VS matrix The 38th entry at all entries for terminating.

In these and other cases, audio coding apparatus 20 is further configured to select the subset of the vector of VS matrix To indicate different audio object.In these and other cases, audio coding apparatus 20 is configured to four of selection VS matrix Vector, and four wherein selected vectors have four maximum directivity quotient in institute's directed quantity of VS matrix.

In these and other cases, audio coding apparatus 20 is configured to determine the vector for indicating different audio object Selection subsets are both directionality and energy based on each vector.

In these and other cases, audio coding apparatus 20, which is further configured to, indicates the one of different audio object Or between multiple primary vectors and one or more secondary vectors execute energy comparison with the rearranged sequence of determination one or more first Vector, wherein one or more described primary vectors describe the different audio object in the first part of audio data, and described one Or multiple secondary vectors describe the different audio object in the second part of audio data.

In these and other cases, audio coding apparatus 20, which is further configured to, indicates the one of different audio object Or between multiple primary vectors and one or more secondary vectors execute crosscorrelation with the rearranged sequence of determination one or more first Vector, wherein one or more described primary vectors describe the different audio object in the first part of audio data, and described one Or multiple secondary vectors describe the different audio object in the second part of audio data.

In some cases, the various aspects of the technology also aloow audio coding apparatus 20 to be configured to relatively Energy compensating is executed in the decomposition of HOA coefficient 11.In these and other cases, audio coding apparatus 20 can be configured with phase Executing the synthesis based on vector for multiple spherical harmonic coefficients indicates one or more audio objects and corresponding direction information to generate The multiple spherical harmonic coefficient through exploded representation, wherein the spherical harmonic coefficient is associated with an order and description sound field;From described Directional information determines different and background directional information;Reduce the order of associated with background audio object directional information with Generate transformed background directional information;Using compensation to increase the value of transformed directional information to save the total energy of the sound field Amount.

In these and other cases, audio coding apparatus 20 can be configured with: execute surprise relative to multiple spherical harmonic coefficients Different value is decomposed to generate the U matrix and s-matrix and the V matrix for indicating directional information that indicate audio object;Determine the phase of V matrix The background column vector of heterotaxy vector sum V matrix;Reduce the order of the background column vector of V matrix to generate the transformed back of V matrix Scape column vector;And application is compensated to increase the value of the transformed background column vector of V matrix to save the total energy of sound field.

In these and other cases, audio coding apparatus 20 is further configured to determine the protrusion singular value of s-matrix Number, wherein the number of the different column vector of V matrix be s-matrix protrusion singular value number.

In these and other cases, audio coding apparatus 20 is configured to determine the order that reduced of spherical harmonic coefficient, and The value of pulverised and the row for being greater than the background column vector through the associated V matrix of the reduction order of order.

In these and other cases, audio coding apparatus 20 is further configured to background column, the S square of combination U matrix The transposition that the background column of battle array and the transformed background of V matrix arrange is to generate modified spherical harmonic coefficient.

In these and other cases, the modified spherical harmonic coefficient describes one or more background components of sound field.

In these and other cases, audio coding apparatus 20 is configured to determine the vector of the background column vector of V matrix The first energy and V matrix transformed background column vector vector the second energy, and value of magnification is applied to the warp of V matrix The each element for converting the vector of background column vector, wherein the value of magnification includes the ratio of the first energy and the second energy Rate.

In these and other cases, audio coding apparatus 20 is configured to determine the vector of the background column vector of V matrix The first root mean square energy and V matrix transformed background column vector vector the second root mean square energy, and value of magnification is answered Each element of the vector of transformed background column vector for V matrix, wherein the value of magnification includes the first energy and second The ratio of energy.

The various aspects of technology described in the present invention also aloow audio coding apparatus 20 relative to HOA coefficient 11 version of being decomposed executes interpolation.In some cases, audio coding apparatus 20 can be configured at least partially through phase The first of spherical harmonic coefficient a for more than first is decomposed and the second decomposition of more than second a spherical harmonic coefficients executes interpolation and obtains for the moment Between segment the interpolated spherical harmonic coefficient of breakdown.

In these and other cases, described first decompose include indicate the right side of more than first a spherical harmonic coefficient it is unusual to First V matrix of amount.

In these and other example, described second decompose include indicate the right side of more than second a spherical harmonic coefficient it is unusual to 2nd V matrix of amount.

In these and other cases, described first decompose include indicate the right side of more than first a spherical harmonic coefficient it is unusual to First V matrix of amount, and described second decomposes the 2nd V square including indicating the right singular vector of more than second a spherical harmonic coefficient Battle array.

In these and other cases, the time slice includes the subframe of audio frame.

In these and other cases, the time slice includes the time samples of audio frame.

In these and other cases, it is humorous to be configured to the ball obtained in more than first a spherical harmonic coefficients for audio coding apparatus 20 The first of coefficient is decomposed and the interpolated decomposition of the second decomposition.

In these and other cases, audio coding apparatus 20 is configured to obtain to contain described in the first frame First of the first part of a spherical harmonic coefficient more than first decomposes and to contain a humorous system of ball more than described second in the second frame The interpolated decomposition that the second of several second parts is decomposed, and be further configured to will be described interpolated for audio coding apparatus 20 The first time component for being applied to the first part comprising a spherical harmonic coefficient more than described first in the first frame is decomposed to generate First manual time's component of a spherical harmonic coefficient more than described first, and corresponding interpolated decompose is applied to comprising in the second frame More than described second a spherical harmonic coefficients second part the second time component it is humorous to generate more than included described second a ball Second manual time's component of coefficient.

In these and other cases, the first time component is by holding relative to more than described first a spherical harmonic coefficients Row the synthesis based on vector and generate.

In these and other cases, the second time component is by holding relative to more than described second a spherical harmonic coefficients Row the synthesis based on vector and generate.

In these and other cases, audio coding apparatus 20 is further configured to: receiving first manual time Component and the second manual time component;Calculate for more than described first a spherical harmonic coefficients first part first decompose and The interpolated decomposition that second of second part for more than described second a spherical harmonic coefficients is decomposed;And by the interpolated decomposition It is inverse to be applied to the first manual time component to restore the first time component and be applied to second manual time point Amount is to restore the second time component.

In these and other cases, audio coding apparatus 20 is configured to of a spherical harmonic coefficient more than first described in interpolation The second space component of one spatial component and more than second a spherical harmonic coefficient.

In these and other cases, first spatial component includes the left surprise for indicating more than first a spherical harmonic coefficient First U matrix of incorgruous amount.

In these and other cases, the second space component includes the left surprise for indicating more than second a spherical harmonic coefficient 2nd U matrix of incorgruous amount.

In these and other cases, first spatial component indicates that the ball for more than described first a spherical harmonic coefficients is humorous M time slice of coefficient, and M of the second space representation in components for the spherical harmonic coefficient of more than described second a spherical harmonic coefficients A time slice.

In these and other cases, first spatial component indicates the spherical harmonic coefficient for being used for more than first a spherical harmonic coefficients M of spherical harmonic coefficient for more than described second a spherical harmonic coefficients of M time slice and the second space representation in components when Between segment, and audio coding apparatus 20 is configured to last N number of element of the first spatial component described in interpolation and described second empty Between component top n element.

In these and other cases, a spherical harmonic coefficient more than described second is in the time domain in more than described first a spherical harmonic coefficients Later.

In these and other cases, audio coding apparatus 20 is further configured to decompose a humorous system of ball more than described first Number is decomposed with generate a spherical harmonic coefficient more than described first first.

In these and other cases, audio coding apparatus 20 is further configured to decompose a humorous system of ball more than described second Number is decomposed with generate a spherical harmonic coefficient more than described second second.

In these and other cases, audio coding apparatus 20 is further configured to humorous relative to more than described first a balls Coefficient executes singular value decomposition with described in generating the U matrix for the left singular vector for indicating a spherical harmonic coefficient more than described first, indicating The V matrix of the s-matrix of the singular value of a spherical harmonic coefficient more than first and the right singular vector of expression more than first a spherical harmonic coefficient.

In these and other cases, audio coding apparatus 20 is further configured to humorous relative to more than described second a balls Coefficient executes singular value decomposition with described in generating the U matrix for the left singular vector for indicating a spherical harmonic coefficient more than described second, indicating The V matrix of the s-matrix of the singular value of a spherical harmonic coefficient more than second and the right singular vector of expression more than second a spherical harmonic coefficient.

In these and other cases, a spherical harmonic coefficient more than described first and second respectively indicates that the plane wave of sound field indicates.

In these and other cases, a spherical harmonic coefficient more than described first and second respectively indicates to mix one or more A monophonic audio object.

In these and other cases, a spherical harmonic coefficient more than described first and second respectively includes the corresponding of expression three-dimensional sound field First and second spherical harmonic coefficients.

In these and other cases, a spherical harmonic coefficient more than described first and second respectively with have greater than one order extremely A few spherical basis function is associated.

In these and other cases, a spherical harmonic coefficient more than described first and second respectively with have equal to four order extremely A few spherical basis function is associated.

In these and other cases, the interior weighted interpolation for being inserted as first decomposition and the second decomposition, wherein The weight for the weighted interpolation decomposed applied to described first and time for being indicated by the described first and second vectors for decomposing at Inverse ratio, and the vector table for being wherein applied to the weight for the weighted interpolation that described second decomposes and being decomposed by described first and second The time shown is proportional.

In these and other cases, the interpolated spherical harmonic coefficient of breakdown smooths more than first a spherical harmonic coefficient and institute State at least one of the spatial component of a spherical harmonic coefficient and time component more than second.

In these and other cases, audio coding apparatus 20 is configured to calculate Us [n]=HOA (n) * (V_vec [n]) -1 to obtain a scalar.

In these and other cases, interpolation includes linear interpolation.In these and other cases, interpolation includes non-linear Interpolation.In these and other cases, the interpolation includes cosine interpolation.In these and other cases, interpolation includes through adding Weigh cosine interpolation.In these and other cases, interpolation includes cubic interpolation.In these and other cases, interpolation includes certainly Adaptability spline interpolation.In these and other cases, interpolation includes minimum curvature interpolation.

In these and other cases, audio coding apparatus 20 is further configured to generate comprising for time slice The bit stream of the instruction of the type of the expression and interpolation of the interpolated spherical harmonic coefficient of breakdown.

In these and other cases, the instruction includes one or more positions for being mapped to the type of interpolation.

By this method, the various aspects of technology described in the present invention aloow audio coding apparatus 20 to be configured With obtain comprising for time slice the interpolated spherical harmonic coefficient of breakdown indicate and the bit stream of the instruction of the type of interpolation.

In these and other cases, the instruction includes one or more positions for being mapped to the type of interpolation.

In this regard, audio coding apparatus 20 can indicate one embodiment of the technology, this is because in some cases Under, audio coding apparatus 20 can be configured to generate the spatial component including sound field through compressed version, by relative to multiple Spherical harmonic coefficient executes the synthesis based on vector and the bit stream of spatial component that generates.

In these and other cases, audio coding apparatus 20 is configured to generate bit stream further to press comprising specifying The field of the prediction mode used when contracting spatial component.

In these and other cases, audio coding apparatus 20 is configured to generate bit stream comprising specifying in compression space The huffman table information of the huffman table used when component.

In these and other cases, audio coding apparatus 20 is configured to generate bit stream to compress comprising instruction expression The field of the value of the quantization step used when spatial component or its parameter.

In these and other cases, described value includes nbits value.

In these and other cases, audio coding apparatus 20 is configured to generate bit stream with (empty comprising it comprising sound field Between component through compressed version) multiple spatial components through compressed version, wherein described value expression is when compressing the multiple sky Between component when the quantization step or its parameter that use.

In these and other cases, audio coding apparatus 20 is further configured to generate bit stream comprising Huffman code To indicate the class identifier of compression classification corresponding to identification spatial component.

In these and other cases, audio coding apparatus 20 is configured to generate bit stream to be comprising identification spatial component The sign bits of positive value or negative value.

In these and other cases, audio coding apparatus 20 be configured to generate bit stream with comprising Huffman code to indicate The residual value of spatial component.

In these and other cases, the synthesis based on vector includes singular value decomposition.

In this regard, audio coding apparatus 20 can further implement the various aspects of the technology, this is because some In the case of, audio coding apparatus 20 can be configured with based on multiple spatial components through in compression space component relative to described The order identification huffman codebook of remaining person in multiple spatial components when compressing the spatial component to use, the space Component is generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

In these and other cases, it is used when audio coding apparatus 20 is configured to based on the compression spatial component Prediction mode identifies huffman codebook.

In these and other cases, at least partly using through compressed version for spatial component identifies huffman codebook Huffman table information and be shown in bit stream.

In these and other cases, at least partly using through compressed version for spatial component indicates described in expression compression The field of the value of the quantization step used when spatial component or its parameter and be shown in bit stream.

In these and other cases, described value includes nbits value.

In these and other cases, bit stream includes multiple spaces of sound field (including its spatial component through compressed version) Component through compressed version, and described value expresses the quantization step or its parameter used when compressing the multiple spatial component.

In these and other cases, spatial component through compressed version at least partly uses Huffman code to select form And be shown in bit stream, the huffman codebook identified is to indicate that the classification of compression classification corresponding to identification spatial component is known It does not accord with.

In these and other cases, spatial component through compressed version at least partly using identification spatial component be positive The sign bits of value or negative value and be shown in bit stream.

In these and other cases, spatial component through compressed version at least partly uses Huffman code to select form And it is shown in bit stream, residual value of the huffman codebook identified to representation space component.

In these and other cases, audio coding apparatus 20 is further configured to based on the huffman codebook identified Compression space component with generate spatial component through compressed version, and generate bit stream with comprising spatial component through compressed version.

In addition, in some cases, the various aspects of the implementable technology of audio coding apparatus 20, this is because audio Code device 20 can be configured with determine compression sound field spatial component (by relative to multiple spherical harmonic coefficients execute based on to The synthesis of amount and the spatial component generated) when the quantization step that will use.

In these and other cases, audio coding apparatus 20 is further configured to be determined based on targeted bit rates and be quantified Step-length.

In these and other cases, audio coding apparatus 20 is configured to determine the number of the position to representation space component Purpose estimation, and the quantization step is determined based on the difference between the estimation and targeted bit rates.

In these and other cases, audio coding apparatus 20 is configured to determine the number of the position to representation space component Purpose estimation determines the difference between the estimation and targeted bit rates, and true and the difference is added to targeted bit rates The fixed quantization step.

In these and other cases, audio coding apparatus 20 is configured to calculate the position that will be generated for spatial component Number estimation (the given codebook corresponding to targeted bit rates).

In these and other cases, sound audio coding apparatus 20 is configured to calculate the position that will be generated for spatial component Number estimation (to the decoding mode used when being scheduled on compression space component).

In these and other cases, audio coding apparatus 20 is configured to: calculating the position that will be generated for spatial component Number first estimation (to the first decoding mode that will be used when being scheduled on compression space component);Calculating will be directed to spatial component Second estimation (to the second decoding mode that will be used when being scheduled on compression space component) of the number of the position of generation;Selection first is estimated The position with minimal number will act as the estimation of the identified estimation of bits number in meter and the second estimation.

In these and other cases, audio coding apparatus 20 is configured to: class corresponding to identification identification spatial component Other class identifier;Identify the residual value for the spatial component that will lead to when compression corresponds to the spatial component of the classification Bit length;And it determines at least partially through the bits number to indicate class identifier is added to the bit length of residual value The estimation of bits number.

In these and other cases, audio coding apparatus 20 is further configured to select to incite somebody to action in compression space component One of multiple codebooks used.

In these and other cases, audio coding apparatus 20 is further configured to using every in the multiple codebook One determines the estimation of the number of the position to representation space component, and selects to cause in the multiple codebook with minimal number Position identified estimation a codebook.

In these and other cases, audio coding apparatus 20 is further configured to using one in the multiple codebook Or more persons determine the estimation of the number of the position to representation space component, the one or more in the multiple codebook is to be based on The order of element number of the spatial component of other elements relative to spatial component compression is selected.

In these and other cases, audio coding apparatus 20 is further configured to using the warp in the multiple codebook Design is determined with the codebook used when spatial component is not to predict from subsequent spatial component to indicate the space point The estimation of the number of the position of amount.

In these and other cases, audio coding apparatus 20 is further configured to using the warp in the multiple codebook Design is determined with the codebook used when spatial component is predicted from subsequent spatial component to indicate the spatial component Position number estimation.

In these and other cases, audio coding apparatus 20 is further configured to using the warp in the multiple codebook Design is determined with the codebook used when indicating the Composite tone object in sound field in spatial component to indicate the space The estimation of the number of the position of component.

In these and other cases, Composite tone object includes through pulse-code modulation (PCM) audio object.

In these and other cases, audio coding apparatus 20 is further configured to using the warp in the multiple codebook Design is determined with the codebook used when spatial component indicates the recorded audio object in sound field to indicate the sky Between component position number estimation.

In each of above-mentioned various examples, it should be appreciated that 20 executing method of audio coding apparatus, or also comprise Execute the device that audio coding apparatus 20 is configured to each step of the method executed.In some cases, these devices can Including one or more processors.In some cases, one or more described processors can be indicated by means of storage to non-transitory The application specific processor of the instruction configuration of computer-readable storage medium.In other words, in each of set of encoding example The various aspects of technology can provide the non-transitory computer-readable storage media for being stored thereon with instruction, described instruction is being held One or more described processors are caused to execute the method that audio coding apparatus 20 has been configured to execute when row.

Fig. 5 is the block diagram that the audio decoding apparatus 24 of Fig. 3 is described in more detail.As Fig. 5 example in show, audio decoder Device 24 may include extraction unit 72, the reconstruction unit 90 based on directionality and the reconstruction unit based on vector 92.

Extraction unit 72 can indicate to be configured to receive bit stream 21 and extract the various encoded version (examples of HOA coefficient 11 Such as, the encoded version based on direction or the encoded version based on vector) unit.Extraction unit 72 can be from marking above Syntactic element (for example, ChannelType syntactic element shown in the example of Figure 10 E and 10H (i) to 10O (ii)) determines HOA Whether coefficient 11 encodes via the various versions.When executing the coding based on direction, extraction unit 72 can extract HOA system (it is expressed as base in the example of fig. 5 for the version based on direction of number 11 and syntactic element associated with this encoded version Information 91 in direction), this information 91 based on direction is transmitted to the reconstruction unit 90 based on direction.This weight based on direction It builds unit 90 and can indicate to be configured to information 91 based on described based on direction and rebuild HOA coefficient in the form of HOA coefficient 11' Unit.The example of bit stream and bit stream inner syntax element being arranged in below with respect to Figure 10 to 10O (ii) and 11 is retouched in more detail It states.

When syntactic element indicates to encode HOA coefficient 11 using the synthesis based on vector, the extractable warp of extraction unit 72 Decoding prospect V [k] vector 57, encoded environment HOA coefficient 59 and encoded nFG signal 59.Extraction unit 72 can will be through decoding Prospect V [k] vector 57 is transmitted to quantifying unit 74, and encoded environment HOA coefficient 59 is provided together with encoded nFG signal 61 To psychologic acoustics decoding unit 80.

To extract through decoding prospect V [k] vector 57, encoded environment HOA coefficient 59 and encoded nFG signal 59, extract Unit 72 can get the side channel information 57 of the syntactic element comprising being expressed as codedVVecLength.Extraction unit 72 can CodedVVecLength is parsed from side channel information 57.Extraction unit 72 can be configured in configuration mould as described above It is operated in any one of formula based on codedVVecLength syntactic element.

Extraction unit 72 is then operated according to any one of configuration mode to parse from side channel information 57 through compressing Form through reduce prospect V [k] vector 55k.Extraction unit 72 can according to presented in following pseudo-code switch narration and with The grammer that is presented in the syntax table of lower VVectorData and operate:

In aforementioned syntax table, there are four types of the first switch of situation (situation 0 to 3) narrations to provide so as to according to coefficient for tool Number (VVecLength) and index (VVecCoeffId) determine VT DISTThe mode of vector length.First situation (situation 0) refers to Show for VT DISTAll coefficients (NumOfHoaCoeffs) of vector are designated.Second situation, situation 1 indicate only VT DISTVector Those of the order coefficient corresponded to greater than MinNumOfCoeffsForAmbHOA be designated, can indicate to be referred to as above (NDIST+1)2-(NBG+1)2The case where.In addition, those of identification in ContAddAmbHoaChan NumOfContAddAmbHoaChan coefficient is subtracted.List ContAddAmbHoaChan is specified to correspond to over order (wherein " channel " refers to the specific system corresponding to a certain order, the combination of sub- rank to the additional channels of the order of MinAmbHoaOrder Number).Third situation, situation 2 indicate VT DISTVector corresponds to that for being greater than the order of MinNumOfCoeffsForAmbHOA A little coefficients are designated, and can indicate to be referred to as (N aboveDIST+1)2-(NBG+1)2The case where.4th situation, situation 3, instruction VT DISTThose of remaining coefficient after removing the coefficient identified by NumOfAddAmbHoaChan of vector is designated. Both VVecLength and VVecCoeffId lists are effective for all VVector on HOAFrame.

After this switch narration, it can control whether to execute by NbitsQ (or, as just indicated, nbits) uniform The decision of de-quantization executes uniform 8 scalar de-quantizations if NbitsQ is equal to 5.With this comparison, it is greater than or equal to 6 NbitsQ value can lead to the application of Hofmann decoding.Cid value mentioned above can be equal to two of NbitsQ value it is minimum effectively Position.Prediction mode discussed herein above is expressed as PFlag in the above syntax table, and HT information bit indicates in the above syntax table For CbFlag.The specified decoding of remaining grammer occurs as being how substantially similar to the mode of mode as described above.Hereafter close The various of the bit stream 21 for meeting each of above-indicated various situations are more fully described in Figure 10 H (i) to 10O (ii) Example.

Reconstruction unit 92 expression based on vector is configured to execute and above for the synthesis unit 27 based on vector It is described to operate reciprocal operation to rebuild the unit of HOA coefficient 11'.Reconstruction unit 92 based on vector may include quantization Unit 74, space-time interpolation unit 76, prospect work out unit 78, psychologic acoustics decoding unit 80, HOA coefficient and work out unit It 82 and reorders unit 84.

Quantifying unit 74 can indicate to be configured and operate in the reciprocal mode of quantifying unit 52 shown in the example with Fig. 4 So as to through decoding 57 de-quantization of prospect V [k] vector and whereby generate through reduce prospect V [k] vector 55kUnit.De-quantization Unit 74 can in some instances by with described above in relation to quantifying unit 52 in a manner of reciprocal mode execute a kind of form Entropy decoding and scalar de-quantization.Dequantizing unit 74 can will through reduce prospect V [k] vector 55kIt is forwarded to the unit 84 that reorders.

Psychologic acoustics decoding unit 80 can psychologic acoustics audio coding unit 40 shown in the example with Fig. 4 it is reciprocal Mode operates to decode encoded environment HOA coefficient 59 and encoded nFG signal 61 and generate the ring through energy compensating whereby Border HOA coefficient 47' and interpolated nFG signal 49'(its be also known as interpolated nFG audio object 49').Psychologic acoustics decoding Environment HOA coefficient 47' through energy compensating can be transmitted to HOA coefficient and work out unit 82 and transmit nFG signal 49' by unit 80 To the device 84 that reorders.

The unit 84 that reorders can indicate to be configured and by with above in relation in a manner of reordering described in unit 34 substantially The unit that reciprocal mode operates.The unit 84 that reorders can receive the language of the original order of the prospect component of instruction HOA coefficient 11 Method element.The unit 84 that reorders can reorder syntactic element to interpolated nFG signal 49' and through reducing prospect V [k] based on these Vector 55kReorder to generate the nFG signal 49 of rearranged sequence " and prospect V [k] vector 55 of rearranged sequencek'.Reorder unit 84 can be by the nFG signal 49 of rearranged sequence " is output to prospect and works out unit 78, and by prospect V [k] vector 55 of rearranged sequencek' It is output to space-time interpolation unit 76.

Space-time interpolation unit 76 can be similar to above in relation to mode described by space-time interpolation unit 50 Mode operate.Space-time interpolation unit 76 can receive prospect V [k] vector 55 of rearranged sequencek'And for rearranged sequence Prospect V [k] vector 55k'With prospect V [k-1] vector 55 of rearranged sequencek-1'Execute space-time interpolation with generate it is interpolated before Scape V [k] vector 55k".Space-time interpolation unit 76 can be by interpolated prospect V [k] vector 55k"It is forwarded to prospect and works out unit 78。

Prospect works out unit 78 and can indicate to be configured to about interpolated prospect V [k] vector 55k"With rearranged sequence NFG signal 49 " executes matrix multiplication to generate the unit of prospect HOA coefficient 65.Prospect, which works out unit 78, can be performed rearranged sequence NFG signal 49 " multiplied by interpolated prospect V [k] vector 55k"Matrix multiplication.

HOA coefficient work out unit 82 can indicate to be configured to be added to prospect HOA coefficient 65 environment HOA channel 47' with Just obtain HOA coefficient 11' unit, wherein apostrophe notation reflect these HOA coefficients 11' can be similar to HOA coefficient 11 but it is non-with Its is identical.Difference between HOA coefficient 11 and 11' can be due to being attributed to the transmitting damaged on transmitting media, quantization or other have The loss that damage operation generates.

By this method, the technology aloows the audio decoding apparatus such as audio decoding apparatus 24: true from bit stream Fixed quantified directional information, encoded prospect audio object and encoded environment high-order ambiophony (HOA) coefficient, wherein through measuring Change directional information and encoded prospect audio Object table shows the prospect HOA coefficient for describing the prospect component of sound field, and wherein warp knit Code environment HOA coefficient describes the context components of sound field;De-quantization is carried out to generate directional information to quantified directional information;Relatively Space-time interpolation is executed in the directional information to generate interpolated directional information;Encoded prospect audio object is carried out Audio decoder is to generate prospect audio object and encoded environment HOA coefficient to generation environment HOA coefficient;As interpolated side Prospect HOA coefficient is determined to the function of information and prospect audio object;And as prospect HOA coefficient and the letter of environment HOA coefficient Number determines HOA coefficient.

By this method, the various aspects of the technology aloow unified audio decoding apparatus 24 in two different decompressions Switch between contracting scheme.In some cases, audio decoding apparatus 24 can be configured based on the spherical harmonic coefficient for indicating sound field It whether is one of to select multiple decompression schemes from the instruction that Composite tone object generates, and use institute through compressed version The selected person stated in multiple decompression schemes decompresses the described through compressed version of the spherical harmonic coefficient.At these and other In the case of, the audio decoding apparatus 24 includes integrated form decoder.

In some cases, audio decoding apparatus 24 can be configured to obtain the indicating whether spherical harmonic coefficient of sound field is from conjunction The instruction generated at audio object.

In these and other cases, audio decoding apparatus 24 is configured to from storage spherical harmonic coefficient through compressed version Bit stream obtains the instruction.

By this method, the various aspects of the technology aloow audio decoding apparatus 24 to obtain the different of description sound field The vector of component and background component.In some cases, audio decoding apparatus 24 can be configured to determine the different of description sound field One or more secondary vectors of the background component of one or more primary vectors of component and the description sound field, it is described one or more Both primary vector and one or more secondary vectors be at least through relative to the multiple spherical harmonic coefficient execute transformation and It generates.

In these and other cases, audio decoding apparatus 24, wherein transformation includes singular value decomposition, the singular value point The S that solution generates and indicates the U matrix of the left singular vector of the multiple spherical harmonic coefficient, indicates the singular value of the multiple spherical harmonic coefficient The V matrix of the right singular vector of matrix and the multiple spherical harmonic coefficient 11 of expression.

In these and other cases, audio decoding apparatus 24, wherein one or more described primary vectors include one or more A U through audio codingDIST*SDISTVector, be before audio coding by by U matrix one or more through audio coding UDISTVector multiplied by s-matrix one or more SDISTVector and generate, and wherein U matrix and s-matrix be at least through relative to The multiple spherical harmonic coefficient executes singular value decomposition and generates.

In these and other cases, audio decoding apparatus 24 be further configured to it is described one or more compile through audio The U of codeDIST*SDISTVector carries out audio decoder to generate one or more through audio coding UDIST*SDISTVector through audio decoder Version.

In these and other cases, audio coding apparatus 24, wherein one or more described primary vectors include one or more A U through audio codingDIST*SDISTVector sum has one or more V of the transposition of V matrixT DISTVector, the UDIST*SDISTTo Amount is before audio coding by by one or more U through audio coding of U matrixDISTVector is multiplied by the one or more of s-matrix A SDISTVector and generate, and wherein U matrix and s-matrix and V matrix are at least through relative to the multiple spherical harmonic coefficient It executes singular value decomposition and generates.

In these and other cases, audio decoding apparatus 24 be further configured to it is described one or more compile through audio The U of codeDIST*SDISTVector carries out audio decoder to generate one or more through audio coding UDIST*SDISTVector through audio decoder Version.

In these and other cases, audio decoding apparatus 24 is further configured to UDIST*SDISTVector multiplied by VT DISTVector is humorous with the ball for restoring the humorous middle distinct components for indicating sound field of the multiple ball.

In these and other cases, audio decoding apparatus 24, wherein one or more described secondary vectors include one or more A U through audio codingBG*SBG*VT BGVector, the UBG*SBG*VT BGVector is before audio coding by being included in U square U in battle arrayBGVector is multiplied by the S being contained in s-matrixBGVector and then multiplied by the V in the transposition for being contained in V matrixT BGVector And it generates, and wherein U matrix, s-matrix and V matrix are individually unusual at least through executing relative to the multiple spherical harmonic coefficient Value is decomposed and is generated.

In these and other cases, audio decoding apparatus 24, wherein one or more described secondary vectors include one or more A U through audio codingBG*SBG*VT BGVector, the UBG*SBG*VT BGVector is before audio coding by being included in U square U in battle arrayBGVector is multiplied by the S being contained in s-matrixBGVector and then multiplied by the V in the transposition for being contained in V matrixT BGVector And generate, wherein s-matrix, U matrix and V matrix be at least through relative to the multiple spherical harmonic coefficient execute singular value decomposition and It generates, and wherein audio decoding apparatus 24 is further configured to one or more described U through audio codingBG*SBG*VT BGTo Amount carries out audio decoder to generate one or more U through audio decoderBG*SBG*VT BGVector.

In these and other cases, audio decoding apparatus 24, wherein one or more described primary vectors are included in audio By by one or more U through audio coding of U matrix before codingDISTVector multiplied by s-matrix one or more SDISTVector and V One or more V of the transposition of matrixT DISTVector and one or more U through audio coding generatedDIST*SDISTVector, wherein U square Battle array, s-matrix and V matrix are generated at least through the singular value decomposition relative to the multiple spherical harmonic coefficient is executed, and its middle pitch Frequency decoding apparatus 24 is further configured to one or more U through audio coding described in audio decoderDIST*SDISTVector is to generate One or more described UDIST*SDISTVector, and by UDIST*SDISTVector is multiplied by VT DISTVector is to restore the multiple spherical harmonic coefficient Those of the distinct components of middle description sound field spherical harmonic coefficient, wherein one or more described secondary vectors are included in front of audio coding just Pass through the U being included in U matrixBGVector is multiplied by the S being contained in s-matrixBGVector and then multiplied by being contained in V matrix V in transpositionT BGVector and one or more U through audio coding generatedBG*SBG*VT BGVector, and wherein audio decoding apparatus 24 are further configured to one or more U through audio coding described in audio decoderBG*SBG*VT BGVector is to restore to describe sound field Background component the multiple spherical harmonic coefficient at least part, and the multiple ball that will describe the distinct components of sound field is humorous Coefficient is added the multiple to generate with described at least part of the multiple spherical harmonic coefficient of the background component of description sound field The reconstructed version of spherical harmonic coefficient.

In these and other cases, audio decoding apparatus 24, wherein one or more described primary vectors include one or more A UDIST*SDISTVector sum has one or more V of the transposition of V matrixT DISTVector, the UDIST*SDISTVector is compiled in audio By by one or more U through audio coding of U matrix before codeDISTVector multiplied by s-matrix one or more SDISTVector and produce It is raw, wherein U matrix, s-matrix and V matrix be at least through for executing singular value decomposition relative to the multiple spherical harmonic coefficient and Generate, and wherein audio decoding apparatus 20 be further configured to obtain instruction will be extracted from bit stream with described in being formed one or more UDIST*SDISTOne or more V described in vector sumT DISTThe value D of the number of vectors of vector.

In these and other cases, audio decoding apparatus 24, wherein one or more described primary vectors include one or more A UDIST*SDISTVector sum has one or more V of the transposition of V matrixT DISTVector, the UDIST*SDISTVector is compiled in audio By by one or more U through audio coding of U matrix before codeDISTVector multiplied by s-matrix one or more SDISTVector and produce It is raw, wherein U matrix, s-matrix and V matrix be at least through for executing singular value decomposition relative to the multiple spherical harmonic coefficient and Generate, and wherein audio decoding apparatus 24 is further configured to based on audio frame acquisition value D one by one, described value D instruction to from Bit stream is extracted to form one or more described UDIST*SDISTOne or more V described in vector sumT DISTThe number of the vector of vector.

In these and other cases, audio decoding apparatus 24, wherein the transformation includes principal component analysis to identify sound The distinct components of field and the background component of sound field.

The various aspects of technology described in the present invention also aloow audio coding apparatus 24 relative to HOA coefficient Through decompose version execute interpolation.In some cases, audio decoding apparatus 24 can be configured at least partially through opposite The second decomposition execution interpolation with more than second a spherical harmonic coefficients is decomposed in the first of more than first a spherical harmonic coefficients and obtains the time The interpolated spherical harmonic coefficient of the breakdown of segment.

In these and other cases, described first decompose include indicate the right side of more than first a spherical harmonic coefficient it is unusual to First V matrix of amount.

In these and other example, described second decompose include indicate the right side of more than second a spherical harmonic coefficient it is unusual to 2nd V matrix of amount.

In these and other cases, described first decompose include indicate the right side of more than first a spherical harmonic coefficient it is unusual to First V matrix of amount, and described second decomposes the 2nd V square including indicating the right singular vector of more than second a spherical harmonic coefficient Battle array.

In these and other cases, the time slice includes the subframe of audio frame.

In these and other cases, the time slice includes the time samples of audio frame.

In these and other cases, it is humorous to be configured to the ball obtained in more than first a spherical harmonic coefficients for audio decoding apparatus 24 The first of coefficient is decomposed and the interpolated decomposition of the second decomposition.

In these and other cases, audio decoding apparatus 24 is configured to obtain to contain described in the first frame First of the first part of a spherical harmonic coefficient more than first decomposes and to contain a humorous system of ball more than described second in the second frame The interpolated decomposition that the second of several second parts is decomposed, and be further configured to will be described interpolated for audio decoding apparatus 24 The first time component for being applied to the first part comprising a spherical harmonic coefficient more than described first in the first frame is decomposed to generate First manual time's component of a spherical harmonic coefficient more than described first, and corresponding interpolated decompose is applied to comprising in the second frame More than described second a spherical harmonic coefficients second part the second time component it is humorous to generate more than included described second a ball Second manual time's component of coefficient.

In these and other cases, the first time component is by holding relative to more than described first a spherical harmonic coefficients Row the synthesis based on vector and generate.

In these and other cases, the second time component is by holding relative to more than described second a spherical harmonic coefficients Row the synthesis based on vector and generate.

In these and other cases, audio decoding apparatus 24 is further configured to: receiving first manual time Component and the second manual time component;Calculate for more than described first a spherical harmonic coefficients first part first decompose and The interpolated decomposition that second of second part for more than described second a spherical harmonic coefficients is decomposed;And by the interpolated decomposition It is inverse to be applied to the first manual time component to restore the first time component and be applied to second manual time point Amount is to restore the second time component.

In these and other cases, audio decoding apparatus 24 is configured to of a spherical harmonic coefficient more than first described in interpolation The second space component of one spatial component and more than second a spherical harmonic coefficient.

In these and other cases, first spatial component includes the left surprise for indicating more than first a spherical harmonic coefficient First U matrix of incorgruous amount.

In these and other cases, the second space component includes the left surprise for indicating more than second a spherical harmonic coefficient 2nd U matrix of incorgruous amount.

In these and other cases, first spatial component indicates that the ball for more than described first a spherical harmonic coefficients is humorous M time slice of coefficient, and M of the second space representation in components for the spherical harmonic coefficient of more than described second a spherical harmonic coefficients A time slice.

In these and other cases, first spatial component indicates that the ball for more than described first a spherical harmonic coefficients is humorous The M of M time slice of coefficient and the second space representation in components for the spherical harmonic coefficient of more than described second a spherical harmonic coefficients A time slice, and audio decoding apparatus 24 is configured to last N number of element and described of the first spatial component described in interpolation The top n element of two spatial components.

In these and other cases, a spherical harmonic coefficient more than described second is in the time domain in more than described first a spherical harmonic coefficients Later.

In these and other cases, audio decoding apparatus 24 is further configured to decompose a humorous system of ball more than described first Number is decomposed with generate a spherical harmonic coefficient more than described first first.

In these and other cases, audio decoding apparatus 24 is further configured to decompose a humorous system of ball more than described second Number is decomposed with generate a spherical harmonic coefficient more than described second second.

In these and other cases, audio decoding apparatus 24 is further configured to humorous relative to more than described first a balls Coefficient executes singular value decomposition with described in generating the U matrix for the left singular vector for indicating a spherical harmonic coefficient more than described first, indicating The V matrix of the s-matrix of the singular value of a spherical harmonic coefficient more than first and the right singular vector of expression more than first a spherical harmonic coefficient.

In these and other cases, audio decoding apparatus 24 is further configured to humorous relative to more than described second a balls Coefficient executes singular value decomposition with described in generating the U matrix for the left singular vector for indicating a spherical harmonic coefficient more than described second, indicating The V matrix of the s-matrix of the singular value of a spherical harmonic coefficient more than second and the right singular vector of expression more than second a spherical harmonic coefficient.

In these and other cases, a spherical harmonic coefficient more than described first and second respectively indicates that the plane wave of sound field indicates.

In these and other cases, a spherical harmonic coefficient more than described first and second respectively indicates to mix one or more A monophonic audio object.

In these and other cases, a spherical harmonic coefficient more than described first and second respectively includes the corresponding of expression three-dimensional sound field First and second spherical harmonic coefficients.

In these and other cases, a spherical harmonic coefficient more than described first and second respectively with have greater than one order extremely A few spherical basis function is associated.

In these and other cases, a spherical harmonic coefficient more than described first and second respectively with have equal to four order extremely A few spherical basis function is associated.

In these and other cases, the interior weighted interpolation for being inserted as first decomposition and the second decomposition, wherein The weight for the weighted interpolation decomposed applied to described first and time for being indicated by the described first and second vectors for decomposing at Inverse ratio, and the vector table for being wherein applied to the weight for the weighted interpolation that described second decomposes and being decomposed by described first and second The time shown is proportional.

In these and other cases, the interpolated spherical harmonic coefficient of breakdown smooths more than first a spherical harmonic coefficient and institute State at least one of the spatial component of a spherical harmonic coefficient and time component more than second.

In these and other cases, audio decoding apparatus 24 is configured to calculate Us [n]=HOA (n) * (V_vec [n]) -1 to obtain a scalar.

In these and other cases, interpolation includes linear interpolation.In these and other cases, interpolation includes non-linear Interpolation.In these and other cases, the interpolation includes cosine interpolation.In these and other cases, interpolation includes through adding Weigh cosine interpolation.In these and other cases, interpolation includes cubic interpolation.In these and other cases, interpolation includes certainly Adaptability spline interpolation.In these and other cases, interpolation includes minimum curvature interpolation.

In these and other cases, audio decoding apparatus 24 is further configured to generate comprising for time slice The bit stream of the instruction of the type of the expression and interpolation of the interpolated spherical harmonic coefficient of breakdown.

In these and other cases, the instruction includes one or more positions for being mapped to the type of interpolation.

In these and other cases, audio decoding apparatus 24 is further configured to obtain comprising for time slice The bit stream of the instruction of the type of the expression and interpolation of the interpolated spherical harmonic coefficient of breakdown.

In these and other cases, the instruction includes one or more positions for being mapped to the type of interpolation.

In some cases, the various aspects of the technology can be further such that audio decoding apparatus 24 can be configured to The bit stream through compressed version of the spatial component comprising sound field is obtained, the spatial component is by relative to multiple spherical harmonic coefficients It executes the synthesis based on vector and generates.

In these and other cases, spatial component is at least partly using specified when compression space point through compressed version The field of the prediction mode used when amount is shown in bit stream.

In these and other cases, spatial component through compressed version be at least partly using huffman table information and It is shown in bit stream, the huffman table information specifies the huffman table used in compression space component.

In these and other cases, at least partly using through compressed version for spatial component indicates described in expression compression The field of the value of the quantization step used when spatial component or its parameter and be shown in bit stream.

In these and other cases, described value includes nbits value.

In these and other cases, bit stream includes multiple spaces of sound field (including its spatial component through compressed version) Component through compressed version, and described value expresses the quantization step or its parameter used when compressing the multiple spatial component.

In these and other cases, spatial component through compressed version is at least partly shown in position using Huffman code In stream, to indicate the class identifier of compression classification corresponding to identification spatial component.

In these and other cases, spatial component through compressed version at least partly using identification spatial component be positive The sign bits of value or negative value and be shown in bit stream.

In these and other cases, spatial component through compressed version is at least partly used to representation space component Residual value Huffman code and be shown in bit stream.

In these and other cases, described device includes audio decoding apparatus.

It is multiple through in compression space component that the various aspects of the technology also aloow audio decoding apparatus 24 to be based on A spatial component through compressed version identify Hough relative to the multiple order through the rest in compression space component For graceful codebook with the use when unziping it to the spatial component described through compressed version, the spatial component is to pass through phase Synthesis based on vector is executed for multiple spherical harmonic coefficients and is generated.

In these and other cases, audio decoding apparatus 24 be configured to obtain comprising sound field spatial component through pressing The bit stream of contracting version, and at least partly using identified huffman codebook decompression spatial component through compressed version to obtain Spatial component.

In these and other cases, spatial component through compressed version at least partly using specifying in compression space point The field of the prediction mode used when amount and be shown in bit stream, and audio decoding apparatus 24 is configured to be based at least partially on Prediction mode decompression spatial component through compressed version to obtain spatial component.

In these and other cases, spatial component through compressed version at least partly using specifying in compression space point The huffman table information of the huffman table used when amount and be shown in bit stream, and audio decoding apparatus 24 is configured at least portion Divide ground based on the huffman table information decompression spatial component through compressed version.

In these and other cases, spatial component is at least partly empty in compression using instruction expression through compressed version Between component when the field of the value of quantization step or its parameter that uses and be shown in bit stream, and audio decoding apparatus 24 is configured Be based at least partially on described value decompression spatial component through compressed version.

In these and other cases, described value includes nbits value.

In these and other cases, bit stream includes multiple spaces of sound field (including its spatial component through compressed version) The quantization step or its parameter of component used through compressed version, described value expression when compressing the multiple spatial component, and Audio decoding apparatus 24 is configured to be based at least partially on the multiple through compressed version of described value decompression spatial component.

In these and other cases, spatial component at least partly used through compressed version to indicate identification space The Huffman code of the class identifier of compression classification corresponding to component and be shown in bit stream, and audio decoding apparatus 24 is through matching Set be based at least partially on Huffman code decompression spatial component through compressed version.

In these and other cases, spatial component through compressed version at least partly using identification spatial component be positive The sign bits of value or negative value and be shown in bit stream, and audio decoding apparatus 24 be configured to be based at least partially on it is described Sign bits decompress spatial component through compressed version.

In these and other cases, spatial component is at least partly used to representation space component through compressed version Residual value Huffman code and be shown in bit stream, and audio decoding apparatus 24 is configured to be based at least partially on and be contained in In the huffman codebook identified Huffman code decompression spatial component through compressed version.

In each of various examples as described above, it should be appreciated that 24 executing method of audio decoding apparatus or Also comprise the device for executing each step that audio decoding apparatus 24 is configured to the method executed.In some cases, this A little devices may include one or more processors.In some cases, one or more described processors can indicate to be arrived by means of storage The application specific processor of the instruction configuration of non-transitory computer-readable storage media.In other words, in the set of encoding example The various aspects of technology in each can provide the non-transitory computer-readable storage media for being stored thereon with instruction, described Instruction causes one or more described processors to execute the method that audio decoding apparatus 24 has been configured to execute when being executed.

Fig. 6 is content analysis unit (such as the content analysis unit shown in the example of Fig. 4 for illustrating audio coding apparatus 26) flow chart of the example operation of the various aspects of technology described in the present invention is executed.

When whether determine the HOA coefficient 11 for indicating sound field is to generate from Composite tone object, content analysis unit 26 can The frame (93) for obtaining HOA coefficient indicates that (that is, N=4) can multiply 1024 for size 25 for quadravalence.Obtaining frame formula HOA coefficient (it can also be denoted herein as frame formula SHC matrix 11, and subsequent frame formula SHC matrix be represented by frame formula SHC matrix 27B, 27C, etc.) after, content analysis unit 26 can exclude the primary vector of frame formula HOA coefficient 11 then to generate through downscaled frame formula HOA coefficient (94).

Content analysis unit 26 can be predicted then from the residual vector through downscaled frame formula HOA coefficient through downscaled frame formula HOA system Several the first non-vanishing vectors (95).After predicting the first non-vanishing vector, content analysis unit 26 can be first based on prediction non- Null vector and actual non-zero vector and obtain an error (96).Once obtaining the error, content analysis unit 26 can be based on The energy balane ratio (97) of practical first non-vanishing vector and the error.Content analysis unit 26 can then compare this ratio with One threshold value (98).When the ratio is without departing from threshold value ("No" 98), content analysis unit 26 can determine that frame formula SHC matrix 11 is It is generated from record, and refers in bit stream and show the correspondence of SHC matrix 11 indicates it is to generate (100,101) from record through decoding.Work as institute When stating ratio more than threshold value ("Yes" 98), content analysis unit 26 can determine that frame formula SHC matrix 11 is produced from Composite tone object It is raw, and refer in bit stream and show the correspondence of SHC matrix 11 indicates it is to generate (102,103) from Composite tone object through decoding.One In a little situations, when frame formula SHC matrix 11 is generated from record, frame formula SHC matrix 11 is transmitted to base by content analysis unit 26 In the synthesis unit 27 (101) of vector.In some cases, interior when frame formula SHC matrix 11 is generated from Composite tone object Hold analytical unit 26 and frame formula SHC matrix 11 is transmitted to the synthesis unit 28 (104) based on direction.

Fig. 7 is to illustrate that audio coding apparatus (such as audio coding apparatus 20 shown in the example of Fig. 4) executes the present invention Described in the synthetic technology based on vector various aspects example operation flow chart.Initially, audio coding apparatus 20 receive HOA coefficient 11 (106).Audio coding apparatus 20 can call LIT unit 30, can relative to HOA coefficient application LIT with Exporting transformed HOA coefficient, (for example, in the case where SVD, transformed HOA coefficient may include US [k] vector 33 and V [k] vector 35)(107)。

Next audio coding apparatus 20 can call parameter calculation unit 32 in the manner described above relative to US [k] vector 33, US [k-1] vector 33, V [k] and/or V [k-1] vector 35 any combination execute analysis as described above with Identify various parameters.That is, parameter calculation unit 32 can be determined at least based on the analysis of transformed HOA coefficient 33/35 One parameter (108).

Audio coding apparatus 20 can then call the unit 34 that reorders, and the unit 34 that reorders will be transformed based on parameter HOA coefficient (again in the context of SVD, can refer to US [k] vector 33 and V [k] vector 35) reorders rearranged to generate Transformed HOA coefficient 33'/35'(of sequence is or, in other words, US [k] vector 33' and V [k] vector 35'), as retouched above State (109).Audio coding apparatus 20 can also call Analysis of The Acoustic Fields unit during any one of the above operation or subsequent operation 44.Analysis of The Acoustic Fields unit 44 can execute sound field relative to HOA coefficient 11 and/or transformed HOA coefficient 33/35 as described above Analyze sum, the background sound field (N to determine prospect channel (nFG) 45BG) order and the channel additional BG HOA to be sent Number (nBGa) and index (i) (it can be collectively expressed as background channel information 43 in the example in figure 4) (110).

Audio coding apparatus 20 can also call Foreground selection unit 48.Foreground selection unit 48 can be based on background channel information 43 determine background or environment HOA coefficient 47 (112).Audio coding apparatus 20 can further call foreground selection unit 36, prospect Selecting unit 36 can be based on nFG 45 (it can indicate one or more indexes for identifying these prospect vectors) rearranged sequence of selection Those of prospect or the distinct components of sound field vector (114) is indicated in US [k] vector 33' and V [k] the vector 35' of rearranged sequence.

Audio coding apparatus 20 can call energy compensating unit 38.Energy compensating unit 38 can be relative to environment HOA coefficient 47 execution energy compensatings are to compensate energy caused by each for being attributed to and removing the channel HOA by Foreground selection unit 48 (114) Loss, and the environment HOA coefficient 47' through energy compensating is generated whereby.

Audio coding apparatus 20 also then calls space-time interpolation unit 50.Space-time interpolation unit 50 can be opposite In rearranged sequence transformed HOA coefficient 33'/35' execute space-time interpolation with obtain interpolated foreground signal 49'(its Can be referred to " interpolated nFG signal 49' " and remaining developing direction information 53, it is also known as " V [k] vector 53 ") (116). Audio coding apparatus 20 then can call coefficient to reduce unit 46.Coefficient reduces unit 46 can be opposite based on background channel information 43 Executing coefficient reduction in remaining prospect V [k] vector 53, through reduction developing direction information 55, (it is also known as through reducing to obtain Prospect V [k] vector 55) (118).

Audio coding apparatus 20 can call quantifying unit 52 then compressed in the manner described above through reducing prospect V [k] vector 55 and generate through decode prospect V [k] vector 57 (120).

Audio coding apparatus 20 can also call psychological acoustic audio translator unit 40.Psychologic acoustics tone decoder unit 40 can carry out psychologic acoustics decoding to each vector of environment HOA coefficient 47' and interpolated nFG signal 49' through energy compensating To generate encoded environment HOA coefficient 59 and encoded nFG signal 61.Audio coding apparatus then can call bitstream producing unit 42.Bitstream producing unit 42 can be based on through decoding developing direction information 57, be decoded environment HOA coefficient 59, through decoding nFG signal 61 and background channel information 43 generate bit stream 21.

Fig. 8 is retouched to illustrate that audio decoding apparatus (such as audio decoding apparatus 24 shown in Fig. 5) executes in the present invention The flow chart of the example operation of the various aspects for the technology stated.Initially, audio decoding apparatus 24 can receive bit stream 21 (130). After receiving bit stream, audio decoding apparatus 24 can call extraction unit 72.Assume that the instruction of bit stream 21 will for discussion purposes The reconstruction based on vector is executed, extraction element 72 can parse this bit stream to retrieve information referred to above, this information is transmitted To the reconstruction unit 92 based on vector.

In other words, extraction unit 72 can be extracted from bit stream 21 in the manner described above through decoding developing direction letter Breath 57 (again, being also known as through decoding prospect V [k] vector 57) is believed through decoding environment HOA coefficient 59 and through decoding prospect Number (its be also known as through decoding prospect nFG signal 59 or through decode prospect audio object 59) (132).

Audio decoding apparatus 24 can further call quantifying unit 74.Quantifying unit 74 can be to through decoding developing direction information 57 carry out entropy decodings and de-quantization to obtain the developing direction information 55 of reductionk(136).Motive that audio decoding apparatus 24 is also adjustable Manage acoustics decoding unit 80.The encoded environment HOA coefficient 59 of 80 decodable code of psychologic acoustics audio coding unit and encoded prospect Signal 61 is to obtain environment HOA coefficient 47' and interpolated foreground signal 49'(138 through energy compensating).Psychologic acoustics decoding Environment HOA coefficient 47' through energy compensating can be transmitted to HOA coefficient and work out unit 82 and transmit nFG signal 49' by unit 80 To the unit 84 that reorders.

The unit 84 that reorders can receive the syntactic element of the original order of the prospect component of instruction HOA coefficient 11.It reorders Unit 84 can reorder syntactic element to interpolated nFG signal 49' and through reducing prospect V [k] vector 55 based on thesekIt reorders To generate the nFG signal 49 of rearranged sequence " and prospect V [k] vector 55 of rearranged sequencek'(140).Reordering unit 84 can will be through " prospect that is output to works out unit 78, and by prospect V [k] vector 55 of rearranged sequence for the nFG signal 49 that reordersk'It is output to sky M- temporal interpolation unit 76.

Next audio decoding apparatus 24 can call space-time interpolation unit 76.Space-time interpolation unit 76 can connect Receive the developing direction information 55 of rearranged sequencek'And the developing direction information 55 about reductionk/55k-1Execute space-time interpolation To generate interpolated developing direction information 55k"(142).Space-time interpolation unit 76 can be by interpolated prospect V [k] vector 55k"It is forwarded to prospect and works out unit 718.

Audio decoding apparatus 24 can call prospect to work out unit 78.Prospect, which works out unit 78, can be performed interpolated foreground signal 49 " with interpolated developing direction information 55k"Matrix multiplication to obtain prospect HOA coefficient 65 (144).Audio decoding apparatus 24 is also HOA coefficient can be called to work out unit 82.HOA coefficient, which works out unit 82, to be added to the channel environment HOA for prospect HOA coefficient 65 47' is to obtain HOA coefficient 11'(146).

Fig. 9 A to 9L is the block diagram of the various aspects of the audio coding apparatus 20 for the example that Fig. 4 is described in more detail.Fig. 9 A The block diagram of the LIT unit 30 of for more details audio coding apparatus 20.As shown in the example of Fig. 9 A, LIT unit 30 can Include multiple and different Linear Invertible Transforms 200 to 200N.LIT unit 30 may include (providing several examples) singular value decomposition (SVD) 200A (" SVD 200A ") is converted, principal component analysis (PCA) converts 200B (" PCA 200B "), Karhunen-Loeve Convert (KLT) 200C (" KLT 200C "), Fast Fourier Transform (FFT) (FFT) 200D (" FFT 200D ") and discrete cosine transform (DCT)200N("DCT 200N").LIT unit 30 can call any one of these Linear Invertible Transforms 200 relative to HOA The corresponding transformation of the application of coefficient 11 and the corresponding transformed HOA coefficient 33/35 of generation.

It is executed although depicted as directly with respect to HOA coefficient 11, but LIT unit 30 can apply Linear Invertible Transforms 200 To the export item of HOA coefficient 11.For example, LIT unit 30 can be relative to the power spectral density matrix derived from the HOA coefficient 11 Using SVD 200.Power spectral density matrix is represented by the matrix multiplication of PSD and the transposition via hoaFrame to hoaFrame And obtain, it is summarized in pseudo-code as follows.HoaFrame notation refers to the frame of HOA coefficient 11.

LIT unit 30 can get S [k] after SVD 200 (svd) is applied to PSD2Matrix (S_squared) and V [k] matrix.S[k]2Matrix can be referred to a square S [k] matrix, therefore (or alternatively, the SVD unit 200, as one of LIT unit 30 A example) square root calculation can be applied to S [k]2Matrix is to obtain S [k] matrix.In some cases, SVD unit 200 can Execute quantization relative to V [k] matrix to obtain quantified V [k] matrix (it is represented by V [k] ' matrix).LIT unit 30 can lead to It crosses and S [k] Matrix Multiplication is obtained into U [k] matrix with quantified V [k] ' matrix to obtain SV [k] ' matrix first.LIT unit 30 connects Get off the pseudo- (pinv) reciprocal and then that HOA coefficient 11 is reciprocal to obtain multiplied by the puppet of SV [k] ' matrix of available SV [k] ' matrix Obtain U [k] matrix.

Afore-mentioned can be indicated by following pseudo-code:

PSD=hoaFrame'*hoaFrame;

[V, S_squared]=svd (PSD, ' econ');

S=sqrt (S_squared);

U=hoaFrame*pinv (S*V');

Pass through the power spectral density (PSD) relative to HOA coefficient rather than coefficient itself and execute SVD, LIT unit 30 can located The computational complexity for executing SVD is possibly reduced in terms of managing device circulation and the one or more of memory space, while being realized identical Source audio code efficiency, as SVD is directly applied to HOA coefficient.That is, above-mentioned PSD type SVD can potentially demand it is less It calculates, because SVD is completed in F*F matrix (the wherein number that F is HOA coefficient).Compared with M*F matrix, wherein M is frame length Degree, i.e. 1024 or more samples.Via being applied to PSD rather than HOA coefficient 11, with O (M*L when being applied to HOA coefficient 112) It compares, the complexity of SVD now can be about O (L3) (wherein O (*) indicates computational complexity common in computer science and technology Big O notation).

The block diagram of the parameter calculation unit 32 of Fig. 9 B for more details audio coding apparatus 20.Parameter calculation unit 32 It may include energy spectrometer unit 202 and crosscorrelation unit 204.Energy spectrometer unit 202 can be relative to US [k] vector 33 and V One or more of [k] vector 35 execute above-mentioned energy spectrometer with generate one be used in present frame (k) or former frame (k-1) or The relevant parameter (R) of more persons, directional characteristic parameterAnd one or more of energy response (e).Equally, intersect phase The above-mentioned crosscorrelation of one or more of US [k] vector 33 and V [k] vector 35 execution can be used for generate by closing unit 204 The relevant parameter (R) of one or more of present frame (k) or former frame (k-1), directional characteristic parameter And energy response One or more of (e).The exportable current frame parameters 37 of parameter calculation unit 32 and previous frame parameter 39.

The block diagram of the unit 34 that reorders of Fig. 9 C for more details audio coding apparatus 20.The unit 34 that reorders includes Parameter evaluation unit 206 and vector reorder unit 208.The expression of parameter evaluation unit 206 is configured to described above Mode assesses previous frame parameter 39 and current frame parameters 37 to generate the unit for resetting sequence index 205.Resetting sequence index 205 includes Identify the vector of vector sum V [k] vector 35 of US [k] vector 33 by how the index of rearranged sequence is (for example, by having identification The index of current vector position pair the first index and identify described pair of the second rope that the rearranged tagmeme of the vector is set The index drawn to).Vector reorder unit 208 expression be configured to according to reset sequence index 205 to US [k] vector 33 and V [k] The unit that vector 35 reorders.US [k] the vector 33' of the exportable rearranged sequence of the unit 34 that the reorders and V [k] of rearranged sequence to 35' is measured, while will also reset sequence index 205 as one or more syntactic elements and being transmitted to bitstream producing unit 42.

The block diagram of the Analysis of The Acoustic Fields unit 44 of Fig. 9 D for more details audio coding apparatus 20.As Fig. 9 D example in Shown, Analysis of The Acoustic Fields unit 44 may include odd value analysis unit 210A, energy spectrometer unit 210B, spatial analysis unit 210C, spatial concealment analytical unit 210D, diffusion analysis unit 210E and Orientation unit 210F.Odd value analysis unit 210A can indicate to be configured to analyze to generate and the diagonal line value (part for forming US [k] vector 33) by making S vector is successively decreased Slope of a curve unit, wherein big singular value indicates prospect or different sound, and low singular value indicates the background point of sound field Amount, as described above.Energy spectrometer unit 210B can indicate determine V [k] matrix 35 energy with being configured to every vector Unit.

Spatial analysis unit 210C can be indicated to be configured to via HOA coefficient 11 being transformed to spatial domain and identification indicates to answer The high-energy regions of the durection component of the sound field of reservation and the unit for executing dimensional energy as described above analysis.Spatial concealment Analytical unit 210D can indicate to be configured and (only spatial concealment analytical unit 210D can recognize to be similar to dimensional energy analysis Pass through the area of space of spatially close higher-energy sound mask) by way of execute spatial concealment analysis unit.Diffusion Analytical unit 210E can indicate to be configured to execute above-mentioned diffusion analysis relative to HOA coefficient 11 to identify the back that can indicate sound field The unit in the propagation energy region of scape component.Orientation unit 210F can indicate to be configured to execute above-indicated direction point The unit of analysis, the Orientation are related to calculating VS [k] vector, and each to each of these VS [k] vectors Mesh carries out squared and summation to identify directionality quotient.Orientation unit 210F can by each of VS [k] vector this Directionality quotient provides to background/foreground (BG/FG) and identifies (ID) unit 212.

Analysis of The Acoustic Fields unit 44 also may include BG/FG ID unit 212, and the BG/FG ID unit 212 can indicate to be configured The sum in prospect channel is determined with any combination of the analysis based on any combination output by analytical unit 210 to 210F (nFG) 45, the order (N of background sound fieldBG) and the number (nBGa) in the channel additional BG HOA and index (i) (its to be sent Can be referred to collectively as background channel information 43 in the example in figure 4) unit.BG/FG ID unit 212 can determine nFG 45 With background channel information 43 to realize targeted bit rates 41.

The block diagram of the foreground selection unit 36 of Fig. 9 E for more details audio coding apparatus 20.Foreground selection unit 36 Unit 214 is dissected comprising vector, which dissects unit 214 and can dissect or otherwise from the US [k] of rearranged sequence Vector 33' and V [k] the vector 35' of rearranged sequence extracts prospect US [k] vector 49 and prospect V identified by nFG syntactic element 45 [k] vector 51k.Vector, which dissects unit 214, can dissect expression by the identification of Analysis of The Acoustic Fields unit 44 and by 45 (its of nFG syntactic element It is also known as prospect channel information 45) the various vectors of the prospect component of specified sound field.As shown in the example of Fig. 9 E, Under some cases, vector, which dissects unit 214, may be selected prospect US [k] vector 49 and prospect V [k] vector 51kIt is interior it is discontinuous to Measure the prospect component to indicate sound field.In addition, in some cases, vector, which dissects unit 214, may be selected prospect US [k] vector 49 And prospect V [k] vector 51kIdentical vector (setting by turn) indicate the prospect component of sound field.

The block diagram of the Foreground selection unit 48 of Fig. 9 F for more details audio coding apparatus 20.Foreground selection unit 48 It can be based on background channel information (for example, background sound field (NBG) and the number (nBGa) in the channel additional BG HOA to be sent and Index (i)) determine background or environment HOA coefficient 47.For example, work as NBGEqual to for the moment, tool is may be selected in Foreground selection unit 48 There is the HOA coefficient 11 of each sample of the audio frame of the order equal to or less than one.In this example, Foreground selection unit 48 can Then the HOA coefficient 11 with the index by indexing one of (i) identification is selected as additional BG HOA coefficient, wherein will be to The nBGa specified in bit stream 21, which is provided, generates unit 42 in place to make audio decoding apparatus (for example, institute in the example of Fig. 5 The audio decoding apparatus 24 of displaying) BG HOA coefficient 47 can be parsed from bit stream 21.Foreground selection unit 48 can be then by environment HOA coefficient 47 is output to energy compensating unit 38.Environment HOA coefficient 47 can have dimension D: M × [(NBG+1)2+nBGa]。

The block diagram of the energy compensating unit 38 of Fig. 9 G for more details audio coding apparatus 20.Energy compensating unit 38 It can indicate to be configured to execute energy compensating relative to environment HOA coefficient 47 to compensate to be attributed to and be removed by Foreground selection unit 48 Each in the channel HOA and the unit of energy loss generated.Energy compensating unit 38 may include energy determination unit 218, energy Measure analytical unit 220 and energy amplifying unit 222.

Energy determination unit 218 can indicate to be configured to identify US [k] the matrix 33' of the rearranged sequence and V of rearranged sequence The unit of the RMS of every a line and/or column on one or more of [k] matrix 35'.Energy determination unit 38 also can recognize selected (it may include nFG signal 49 and prospect V [k] vector 51 in one or more of prospect channelkAnd the environment HOA of Degree reduction Coefficient 47) every a line and/or column RMS.In US [k] the matrix 33' of rearranged sequence and V [k] the matrix 35' of rearranged sequence Every a line of the one or more and/or the RMS of column can be stored as being expressed as RMSFULLVector, and nFG signal 49, prospect V [k] vector 51kTable can be stored as with every a line of one or more of the environment HOA coefficient 47 of Degree reduction and/or the RMS of column It is shown as RMSREDUCEDVector.

In some instances, believe for V [k] matrix 35', nFG of US [k] the matrix 33' of the rearranged sequence of determination, rearranged sequence Number 49, prospect V [k] vector 51kAnd one or more of environment HOA coefficient 47 of Degree reduction corresponding line and/or column it is every One RMS, energy determination unit 218 will can be applied to the column with reference to spherical harmonic coefficient (SHC) reconstructor first.It is determined by energy single 218 application of member allows to determine the RMS in the domain SHC with US [k] the matrix 33' of the rearranged sequence of determination, rearranged with reference to SHC reconstructor V [k] matrix 35', nFG signal 49, prospect V [k] vector 51 of sequencekWith it is one or more in the environment HOA coefficient 47 of Degree reduction The energy of the described overall sound field of every a line and/or column for the frame that the row and/or column of person indicates.Energy determination unit 38 can incite somebody to action This RMSFULLAnd RMSREDUCEDVector is transmitted to energy spectrometer unit 220.

Energy spectrometer unit 220 can indicate to be configured to the unit that value of magnification vector Z is calculated according to following equation: Z= RMSFULL/RMSREDUCED.This value of magnification vector Z then can be transmitted to energy amplifying unit 222 by energy spectrometer unit 220.Energy Amplifying unit 222 can indicate to be configured to by this value of magnification vector Z or its various certain applications to nFG signal 49, prospect V [k] Vector 51kWith the unit of one or more of the environment HOA coefficient 47 of Degree reduction.In some cases, value of magnification vector Z according to According to following equation HOABG-RED'=HOABG-REDZTIt is applied to the environment HOA coefficient 47 of only Degree reduction, wherein HOABG-REDIndicate rank The environment HOA coefficient 47, HOA of number reductionBG-RED'Indicate the HOA coefficient 47' of the Degree reduction through energy compensating, and ZTIndicate Z-direction The transposition of amount.

The space-time interpolation unit of audio coding apparatus 20 shown in the example of Fig. 9 H for more details Fig. 4 50 block diagram.Space-time interpolation unit 50 can indicate prospect V [k] vector 51 for being configured to receive kth framekAnd former frame Prospect V [k-1] vector 51 of (therefore being k-1 notation)k-1And execute space-time interpolation with generate interpolated prospect V [k] to The unit of amount.Space-time interpolation unit 50 may include V interpolation unit 224 and prospect adaptation unit 226.

V interpolation unit 224 can be based on current prospect V [k] vector 51kAnd previous prospect V [k-1] vector 51k-1Remainder Current prospect V [k] vector 51 of component selectionskThe part for wanting interpolation.The part can be selected as the above finger by V interpolation unit 224 One or more of subframe out or be only may be change from frame to frame single do not define part.In some cases, V interpolation unit 224 may be selected current prospect V [k] vector 51k1024 samples single 128 sample portions with interpolation.V interpolation unit 224 can be then by by current prospect V [k] vector 51kAnd previous prospect V [k-1] vector 51k-1In vector projection to spherical surface (using the projection matrix of such as T design matrix) is gone up each of described vector is converted to independent space reflection.V interpolation Vector in V can be then interpreted as the shape on spherical surface by unit 224.For the V matrix of 256 sample portion of interpolation, V interpolation unit 224 can then these spatial forms of interpolation, and be then transformed back to the humorous domain vector of ball via the inverse of projection matrix.With this side Formula, technology of the invention can provide the smooth transition between V matrix.V interpolation unit 224 can then generate remaining V [k] vector 53, it indicates modified to remove prospect V [k] vector 51kInterpolated part after prospect V [k] vector 51k.V interpolation list Member 224 can be then by interpolated prospect V [k] vector 51k'It is transmitted to nFG adaptation unit 226.

When selecting single part to carry out interpolation, V interpolation unit 224, which can produce, to be expressed as The syntactic element of CodedSpatialInterpolationTime 254, recognition duration or in other words, interpolation when Between (for example, according to the number of sample).When selecting single part to execute subframe interpolation, V interpolation unit 224 also can produce expression For another syntactic element of SpatialInterpolationMethod 255, can recognize performed interpolation type (or In some cases, if execute interpolation).These syntactic elements 254 and 255 can be output to by space-time interpolation unit 50 Bitstream producing unit 42.

NFG adaptation unit 226 can indicate the unit for being configured to generate adapted nFG signal 49'.NFG adaptation unit 226 It can be by first via nFG signal 49 and prospect V [k] vector 51kMultiplication obtain prospect HOA coefficient and generate adapted nFG Signal 49'.After obtaining prospect HOA coefficient, nFG adaptation unit 226 can be by prospect HOA coefficient divided by interpolated prospect V [k] To obtain adapted nFG signal 49'(, it can be referred to interpolated nFG signal 49' to vector 53, it is assumed that these signals are out of warp Insert prospect V [k] vector 51k'Export).

The frame of the coefficient reduction unit 46 of audio coding apparatus 20 shown in the example of Fig. 9 I for more details Fig. 4 Figure.Coefficient is reduced unit 46 and can indicate to be configured to be executed based on background channel information 43 relative to remaining prospect V [k] vector 53 Coefficient is reduced so that reduced prospect V [k] vector 55 to be output to the unit of quantifying unit 52.Reduced prospect V [k] vector 55 can have dimension D:[(N+1)2-(NBG+1)2-nBGa]×nFG。

Coefficient reduction unit 46 may include that coefficient minimizes unit 228, and coefficient minimizes unit 228 and can indicate to be configured to Reduced or by removing any coefficient considered in background HOA coefficient 47 (as identified by background channel information 43) with it Its mode minimizes the unit of the size of each of remaining prospect V [k] vector 53.Coefficient minimizes unit 228 and can be removed Coefficient those of is identified by background channel information 43 to obtain through reducing prospect V [k] vector 55.

The psychologic acoustics tone decoder of audio coding apparatus 20 shown in the example of Fig. 9 J for more details Fig. 4 The block diagram of unit 40.Psychologic acoustics tone decoder unit 40 can indicate to be configured to relative to the background HOA through energy compensating Coefficient 47' and interpolated nFG signal 49' executes the unit of psychologic acoustics coding.As shown in the example of Fig. 9 H, psychologic acoustics sound Frequency translator unit 40 can call multiple items of psychological acoustic audio encoder 40A to 40N to the background through energy compensating (wherein in this context, channel refers to corresponding to specific order/sub- rank spherical substrate function frame in the channel of HOA coefficient 47' In all samples) each of and interpolated nFG signal 49' each signal carry out audio coding.In some instances, Psychologic acoustics tone decoder unit 40 is executed individuation or in other ways comprising (when being implemented with hardware) number enough With each channel to the background HOA coefficient 47' through energy compensating, (or nBGa indexes (i) to audio coder 40A to 40N Sum) and interpolated nFG signal 49'(or nFG) each signal individually encoded and (additional environment added for nBGa in total The sum of the index (i) in the channel HOA adds nFG).The exportable encoded background HOA coefficient 59 of audio coder 40A to 40N and warp Encode nFG signal 61.

The block diagram of the quantifying unit 52 of audio coding apparatus 20 shown in the example of Fig. 9 K for more details Fig. 4. In the example of Fig. 9 K, quantifying unit 52 includes uniform quantifying unit 230, nbits unit 232, predicting unit 234, prediction mould Formula unit 236 (" Pred mode unit 236 "), classification and remaining decoding unit 238 and huffman table selecting unit 240.? The expression of one quantifying unit 230 is configured to that (it can indicate to be reduced prospect V [k] vector 55 relative to one of spatial component Any one of) execute the unit of above-described uniform quantization.The expression of nbits unit 232 is configured to determine nbits parameter Or the unit of value.

Predicting unit 234 indicates to be configured to the unit that prediction is executed relative to quantified spatial component.Predicting unit 234 It can be by with through the upper posterior corresponding person of the time in reduction prospect V [k] vector 55, (it is represented by through reducing prospect V [k- 1] vector 55) it by element subtracts and works as the former through one in reduction prospect V [k] vector 55 and execute prediction.The result of this prediction It can be referred to predicted spatial component.

Prediction mode unit 236 can indicate the unit for being configured to selection prediction mode.Huffman table selecting unit 240 can Indicate the unit for being configured to select decoding of the appropriate huffman table for cid.As an example, prediction mode unit 236 And huffman table selecting unit 240 can be operated according to following pseudo-code:

For giving nbits, all huffman tables with nbits are retrieved

B00=0;B01=0;B10=0;B11=0;// initialize to calculate the expection position of every decoding mode

Element number in for m=1:(# vector)

// calculate the expection number for being used for the position of vector element v (m)

// without prediction and use huffman table 5

B00=B00+calculate_bits (v (m), HT5);

// without prediction and use huffman table { 1,2,3 }

B01=B01+calculate_bits (v (m), HTq);Q in { 1,2,3 }

// calculate the expection number for being used for the position of prediction residue e (m)

E (m)=v (m)-vp (m);//vp (m): previous frame vector element

// have prediction and use huffman table 4

B10=B10+calculate_bits (e (m), HT4);

// have prediction and use huffman table 5

B11=B11+calculate_bits (e (m), HT5);

end

// find out the optimum prediction mode and huffman table for generating minimum position

// optimum prediction mode and huffman table add flag by pflag and Htflag respectively

[Be, id]=min ([B00B01B10B11]);

Switch id

Case 1:pflag=0;HTflag=0;

Case 2:pflag=0;HTflag=1;

Case 3:pflag=1;HTflag=0;

Case 4:pflag=1;HTflag=1;

end

Classification and remaining decoding unit 238 can indicate to be configured to execute in the way of being more fully described above predicted Spatial component or quantified spatial component (when prediction deactivates) classification and remaining decoding unit.

As shown in the example of Fig. 9 K, the exportable various parameters of quantifying unit 52 or value are to contain in bit stream 21 or side In information (it can itself be the bit stream separated with bit stream 21).It is assumed that information is specified in side channel information, then scalar/entropy Quantifying unit 50 can be by nbits value (as nbits value 233), prediction mode (as prediction mode 237) and huffman table information (as huffman table information 241) (is shown as being decoded prospect V in the example in figure 4 together with spatial component through compressed version [k] vector 57) (it in this example, can refer to the Hough for being selected to encode cid, sign bits and block decoding remnants Graceful code) it is output to bitstream producing unit 42.Nbits value can be decoded prospect V [k] vector 57 in side channel information for all In specify primary, and prediction mode and huffman table information can be referred to for through each of decoding prospect V [k] vector 57 It is fixed.Show the part through compressed version of the designated space component of bit stream in more detail in the example of Figure 10 B and/or Figure 10 C.

The frame of the bitstream producing unit 42 of audio coding apparatus 20 shown in the example of Fig. 9 L for more details Fig. 4 Figure.Bitstream producing unit 42 may include that main channel information creating unit 242 and side channel information generate unit 244.Main channel Information creating unit 242 can produce comprising resetting sequence index 205, CodedSpatialInterpolationTime syntactic element 254, SpatialInterpolationMethod syntactic element 255, encoded background HOA coefficient 59 and encoded nFG signal The status of a sovereign stream 21 of one or more of 61 (and if not all).Side channel information generates unit 244 and can indicate to be configured to Generation may include nbits value 233, prediction mode 237, huffman table information 241 and through decoding prospect V [k] vector 57 in one Or the unit of the side channel bit stream 21B of more persons' (and if not all).Bit stream 21 and 21B can be collectively referred to as bit stream 21.Some In context, bit stream 21 can only refer to main channel bitstream 21, and bit stream 21B can be referred to side channel information 21B.

Figure 10 A to 10O (ii) for more details may specify bit stream or side channel information through compression space component Part figure.In the example of Figure 10 A, part 250 include transfer interpreter identifier (" transfer interpreter ID ") field 251 and HOADecoderConfig field 252.Transfer interpreter id field 251 can indicate that storage has been used for the transfer interpreter of the audio mixing of HOA content ID field.HOADecoderConfig field 252 can indicate to be configured to store to initialize HOA spatial decoder The field of information.

HOADecoderConfig field 252 further include directional information (" directional information ") field 253, CodedSpatialInterpolationTime field 254, SpatialInterpolationMethod field 255, CodedVVecLength field 256 and gain information field 257.Directional information field 253 can indicate that storage is based on for configuring The field of the information of the synthesis decoder in direction.CodedSpatialInterpolationTime field 254 can indicate storage base In the field of the time of the space-time interpolation of the signal of vector.SpatialInterpolationMethod field 255 can table Show the field of the instruction for the type of interpolation applied during the space-time interpolation for being stored in the signal based on vector. CodedVVecLength field 256 can indicate to store the length of emitted data vector to synthesize the signal based on vector Field.Gain information field 257 indicates that storage instruction is applied to the field of the information of the gain calibration of signal.

In the example of Figure 10 B, part 258A indicates a part of side information channel, and part of 258A includes frame mark First 259, frame header 259 includes byte number field 260 and nbits field 261.Byte number field 260 can be indicated to table Up to the byte number being contained in the frame for designated space component v1 to vn (comprising zero for byte align field 264) Field.Nbits field 261 indicates to may specify the identified field for for decompressing the nbits value of spatial component v1 to vn.

As further shown in the example of Figure 10 B, part 258A may include the sub- bit stream for v1 to vn, each of these Person is comprising prediction mode field 262, huffman table information field 263 and through the corresponding person in compression space component v1 to vn. Prediction mode field 262 can indicate whether pre- for executing through the corresponding person in compression space component v1 to vn to store The field of the instruction of survey.Huffman table information field 263 is indicated at least partly to indicate to be come using which huffman table The field that various aspects through the corresponding person in compression space component v1 to vn are decoded.

In this regard, the technology aloow audio coding apparatus 20 obtain the spatial component comprising sound field through pressing The bit stream of contracting version, the spatial component are generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

Figure 10 C for more details may specify the substitution of the part 258B of the side channel information through compression space component The figure of example.In the example of Figure 10 C, part 258B includes the frame header 259 comprising Nbits field 261.Nbits field 261 It indicates to may specify the identified field for for decompressing the nbits value of spatial component v1 to vn.

As further shown in the example of Figure 10 C, part 258B may include the sub- bit stream for v1 to vn, each of these Person is comprising prediction mode field 262, huffman table information field 263 and through the corresponding person in compression space component v1 to vn. Prediction mode field 262 can indicate whether pre- for executing through the corresponding person in compression space component v1 to vn to store The field of the instruction of survey.Huffman table information field 263 is indicated at least partly to indicate to be come using which huffman table The field that various aspects through the corresponding person in compression space component v1 to vn are decoded.

In the illustrated example, Nbits field 261 includes subfield A 265, B 266 and C 267.In this example, A 265 and B 266 is respectively 1 bit sub-fields, and C 267 is 2 bit sub-fields.Other examples may include subfield of different sizes 265,266 and 267.A field 265 and B field 266 can indicate first and second most significant bit of storage Nbits field 261 Field, and C field 267 can indicate the field of the least significant bit of storage Nbits field 261.

Part 258B also may include AddAmbHoaInfoChannel field 268.AddAmbHoaInfoChannel field 268 can indicate field of the storage for the information of additional environment HOA coefficient.As shown in the example of Figure 10 C, AddAmbHoaInfoChannel 268 includes CodedAmbCoeffIdx field 246, AmbCoeffIdxTransition field 247.CodedAmbCoeffIdx field 246 can indicate to store the field of the index of additional environment HOA coefficient. AmbCoeffIdxTransition field 247 can indicate to be configured to whether storage instruction has additional environment HOA system in this frame The field of several data for being faded in or being faded out.

Figure 10 C (i) for more details may specify the part 258B''s of the side channel information through compression space component The figure of alternate example.In the example of Figure 10 C (i), part 258B' includes the frame header 259 comprising Nbits field 261. Nbits field 261 indicates to may specify the identified field for for decompressing the nbits value of spatial component v1 to vn.

As further shown in the example of Figure 10 C (i), part 258B' may include the sub- bit stream for v1 to vn, therein Each includes huffman table information field 263 and through the corresponding person in compression direction component v1 to vn, and does not include prediction Pattern field 262.At all other aspect, part 258B' can be similar to part 258B.

The figure of the part 258C of Figure 10 D for more details bit stream 21.Part 258C is similar to part 258, but following Except situation: frame header 259 and zero byte alignment 264 have been removed, while 261 field of Nbits has been added for v1 to vn Each of bit stream before, as shown in the example of Figure 10 D.

The figure of the part 258C' of Figure 10 D (i) for more details bit stream 21.Part 258C' is similar to part 258C, But part 258C' does not include except the prediction mode field 262 for each of V vector v 1 to vn.

The figure of the part 258D of Figure 10 E for more details bit stream 21.Part 258D is similar to part 258B, but following Except situation: frame header 259 and zero byte alignment 264 have been removed, while 261 field of Nbits has been added for v1 to vn Each of bit stream before, as shown in the example of Figure 10 E.

The figure of the part 258D' of Figure 10 E (i) for more details bit stream 21.Part 258D' is similar to part 258D, But part 258D' does not include except the prediction mode field 262 for each of V vector v 1 to vn.In this regard, audio Code device 20 can produce the bit stream 21 not included for each through compressing the prediction mode field 262 of V vector, such as about figure Indicated by the example of 10C (i), 10D (i) and 10E (i).

Figure 10 F is the figure of the part 250 of bit stream 21 shown in the example of explanatory diagram 10A in different ways.The reality of Figure 10 D Example shown in part 250 include HOAOrder field (its be easy illustrate purpose and do not shown in the example of Figure 10 F), MinAmbHoaOrder field (it is to be easy the purpose of explanation and do not show in the example of Figure 10 F again), directional information word Section 253, CodedSpatialInterpolationTime field 254, SpatialInterpolationMethod field 255, CodedVVecLength field 256 and gain information field 257.As shown in the example of Figure 10 F, CodedSpatialInterpolationTime field 254 may include three bit fields, SpatialInterpolationMethod Field 255 may include a bit field, and CodedVVecLength field 256 may include two bit fields.

The figure of the part 248 of Figure 10 G for more details bit stream 21.Part 248 indicates unified voice/tone decoder (USAC) three-dimensional (3D) payload, it includes HOAframe field 249, (it is also denoted as sideband information, side channel letter Breath or side channel bit stream).As shown in the example of Figure 10 E, the expanded view of HOAFrame field 249 can be similar to Figure 10 C's The part 258B of bit stream 21 shown in example." ChannelSideInfoData " includes that (it goes out ChannelType field 269 Do not shown in the example of Figure 10 C in the purpose for being easy explanation), the A field 265 that is expressed as in the example of Figure 10 E " ba ", It is expressed as the B field 266 of " bb " in the example of Figure 10 E and is expressed as the C field 267 of " unitC " in the example of Figure 10 E. ChannelType field indicates that channel is the signal based on direction, the signal based on vector or additional environment HOA coefficient.? Between different ChannelSideInfoData, there are AddAmbHoaInfoChannel fields 268, wherein difference V vector bit stream (for example, " bit stream for v1 " and " bit stream for v2 ") is indicated with grey.

Another various instance section 248H to 248O of Figure 10 H to 10O (ii) for more details bit stream 21 are together with companion With the part HOAconfig 250H to 250O figure.Figure 10 H (i) and 10H (ii) illustrate to have generated the first example bit stream 248H and Adjoint HOA config part 250H is with corresponding with the situation 0 in the above pseudo-code.In the example of Figure 10 H (i), The part HOAconfig 250H includes CodedVVecLength syntactic element 256, is set to indicate that all elements of V vector Through decoding, for example, all 16 V vector elements.The part HOAconfig 250H also includes SpatialInterpolationMethod syntactic element 255 is set to indicate that the interpolating function of space-time interpolation is Raised cosine.In addition, the part HOAconfig 250H includes CodedSpatialInterpolationTime 254, it is set to It is designated as 256 interpolated sample duration.The part HOAconfig 250H further comprises MinAmbHoaOrder grammer member Element 150 is set to indicate that the MinimumHOA order of environment HOA content is one, and wherein audio decoding apparatus 24 can export MinNumofCoeffsForAmbHOA syntactic element is equal to (1+1)2Or four.The part HOAconfig 250H includes HoaOrder language Method element 152 is set to indicate that the HOA order of content is equal to three (or in other words N=3), wherein audio decoding apparatus 24 can export NumOfHoaCoeffs equal to (N+1)2Or 16.

As further shown in the example of Figure 10 H (i), part 248H includes that unified voice and audio coding (USAC) are three-dimensional (USAC-3D) audio frame, two of them HOA frame 249A and 249B are stored in a USAC payload, it is assumed that are enabling frequency spectrum Described two audio frames are stored in a USAC-3D frame when tape copy (SBR).Audio decoding apparatus 24 can export conduct The function of numHOATransportChannels syntactic element and MinNumOfCoeffsForAmbHOA syntactic element it is several Flexible transmission channel.In the following example, assume that numHOATransportChannels syntactic element be equal to 7 and MinNumOfCoeffsForAmbHOA syntactic element is equal to four, and wherein the number in flexible transmission channel is equal to NumHOATransportChannels syntactic element subtracts MinNumOfCoeffsForAmbHOA syntactic element (or three).

Frame 249A and 249B is described in more detail in Figure 10 H (ii).As shown in the example of Figure 10 H (ii), frame 249A includes ChannelSideInfoData (CSID) field 154 to 154C, HOAGainCorrectionData (HOAGCD) field, VVectorData field 156 and 156B and HOAPredictionInfo field.CSID field 154 includes unitC 267, bb 266 and ba265 is set to respective value shown in the example of Figure 10 H (i) together with ChannelType 269, each of these person 01,1,0 and 01.CSID field 154B includes unitC 267, bb 266 and ba 265 together with ChannelType 269, therein Each is set to respective value 01,1,0 and 01 shown in the example of Figure 10 H (ii).CSID field 154C includes with value 3 ChannelType field 269.CSID field 154 is corresponding corresponding to one in transmission channel 1,2 and 3 to each of 154C Person.In effect, every CSID field 154 to 154C indicate corresponding payload 156 and 156B be based on direction signal (when When corresponding ChannelType is equal to zero), signal (when corresponding ChannelType is equal to for the moment), additional environment based on vector HOA coefficient (when corresponding ChannelType is equal to two), or it is empty (when ChannelType is equal to three).

In the example of Figure 10 H (ii), frame 249A includes two signals based on vector (it is assumed that ChannelType 269 It is equal to 1) in CSID field 154 and 154B and a spacing wave is (it is assumed that ChannelType 269 is equal in CSID field 154C 3).The aforementioned part HOAconfig 250H is given, audio decoding apparatus 24 can determine that all 16 V vector elements are encoded.Cause This, VVectorData 156 and 156B respectively contain all 16 vector elements, and each of these person passes through 8 position even amounts Change.As pointed by footnote 1, is decoded the number of VVectorData element and index and referred to by parameter CodedVVecLength=0 It is fixed.In addition, such as pointed by single asterisk (*), decoding scheme is by NbitsQ=5 in the CSID field for corresponding to transmission channel In signal.

In frame 249B, CSID field 154 and 154B in frame 249 CSID field 154 and 154B it is identical, and frame 249B CSID field 154C be switched to the ChannelType for one.Therefore, the CSID field 154C of frame 249B includes Cbflag 267, Pflag 267 (instruction huffman coding) and Nbits 261 (being equal to 12).As a result, frame 249B includes third VVectorData field 156C, it includes 16 V vector elements, and each of these person is by 12 position uniform quantizations and through suddenly Fu Man decoding.As noted above, through decoding the number of VVectorData element and index by parameter CodedVVecLength =0 is specified, and Hoffman decodeng scheme is led to by NbitsQ=12, CbFlag=0 and Pflag=0 for this specific transmission It is signaled in the CSID field 154C in road (for example, the 3rd transmission channel).

The example of Figure 10 I (i) and 10I (ii) illustrate to have generated the second example bit stream 248I and the adjoint portion HOA config Divide 250I with corresponding with the above situation 0 in the above pseudo-code.In the example of Figure 10 I (i), the part HOAconfig 250I packet Syntactic element containing CodedVVecLength 256 is set to indicate that all elements of V vector through decoding, for example, all 16 A V vector element.The part HOAconfig 250I also includes SpatialInterpolationMethod syntactic element 255, warp The interpolating function for being set to indicate that space-time interpolation is raised cosine.In addition, the part HOAconfig 250I includes CodedSpatialInterpolationTime 254 is set to indicate that the interpolated sample duration for 256.

The part HOAconfig 250I further includes MinAmbHoaOrder syntactic element 150, is set to indicate that ring The MinimumHOA order of border HOA content is one, and wherein audio decoding apparatus 24 can export MinNumofCoeffsForAmbHOA Syntactic element is equal to (1+1)2Or four.Audio decoding apparatus 24 can also export MaxNoOfAddActiveAmbCoeffs grammer member Element, the difference being such as set between NumOfHoaCoeff syntactic element and MinNumOfCoeffsForAmbHOA, in this example It is assumed that it is equal to 16-4 or 12.Audio decoding apparatus 24 can also export AmbAsignmBits syntactic element, such as be set to ceil (log2 (MaxNoOfAddActiveAmbCoeffs))=ceil (log2 (12))=4.The part HOAconfig 250H includes HoaOrder syntactic element 152 is set to indicate that the HOA order of content is equal to three (or in other words N=3), middle pitch Frequency decoding apparatus 24 can export NumOfHoaCoeffs equal to (N+1)2Or 16.

As further shown in the example of Figure 10 I (i), part 248H includes USAC-3D audio frame, two of them HOA frame 249C and 249D is stored in the load of USAC extremely efficient, it is assumed that when enabling frequency spectrum tape copy (SBR), described two audio frames It is stored in a USAC-3D frame.Audio decoding apparatus 24 can be exported as numHOATransportChannels syntactic element And several flexible transmission channels of the function of MinNumOfCoeffsForAmbHOA syntactic element.In the following example, it is assumed that NumHOATransportChannels syntactic element is equal to 7 and MinNumOfCoeffsForAmbHOA syntactic element and is equal to four, Wherein the number in flexible transmission channel is equal to numHOATransportChannels syntactic element and subtracts MinNumOfCoeffsForAmbHOA syntactic element (or three).

Frame 249C and 249D is described in more detail in Figure 10 I (ii).As shown in the example of Figure 10 I (ii), frame 249C includes CSID field 154 arrives 154C and VVectorData field 156.CSID field 154 include CodedAmbCoeffIdx 246, (wherein double asterisk (* *) instruction is for the 1st flexible transmission channel, it is herein assumed that decoding by AmbCoeffIdxTransition 247 The internal state of device is AmbCoeffIdxTransitionState=2, causes CodedAmbCoeffIdx bit field in place Signal in stream or specify in other ways) and ChannelType 269 (its be equal to two, to signal correspondence Payload is additional environment HOA coefficient).In this example, AmbCoeffIdx can be exported as and is equal to by audio decoding apparatus 24 CodedAmbCoeffIdx+1+MinNumOfCoeffsForAmbHOA or 5.CSID field 154B includes unitC 267, bb Together with ChannelType 269, each of these person is set to corresponding shown in the example of Figure 10 I (ii) 266 and ba 265 Value 01,1,0 and 01.CSID field 154C includes the ChannelType field 269 with value 3.

In the example of Figure 10 I (ii), frame 249C includes the single signal based on vector (it is assumed that ChannelType 269 1) and a spacing wave (it is assumed that ChannelType 269 in CSID field 154C be equal to 3) it is equal in CSID field 154B.It gives The fixed aforementioned part HOAconfig 250I, audio decoding apparatus 24 can determine that all 16 V vector elements are encoded.Therefore, VVectorData 156 includes all 16 vector elements, and each of these person passes through 8 position uniform quantizations.As footnote 1 is signified Out, the number of VVectorData element and index is decoded to be specified by parameter CodedVVecLength=0.In addition, such as by foot Pointed by note 2, decoding scheme is signaled in the CSID field for corresponding to transmission channel by NbitsQ=5.

In frame 249D, CSID field 154 includes the AmbCoeffIdxTransition that instruction has occurred without transformation 247, and therefore can from former frame imply without signal or in other ways again specify CodedAmbCoeffIdx 246.The CSID field 154B and 154C of frame 249D is identical as the CSID field of frame 249C, and thus, such as 249C at same frame, frame 249D includes single VVectorData field 156, and it includes all 16 vector elements, each of these person is equal by 8 positions Even quantization.

Figure 10 J (i) and 10J (ii) illustrates to have generated the first example bit stream 248J and adjoint HOA config part 250J With corresponding with the situation 1 in the above pseudo-code.In the example of Figure 10 J (i), the part HOAconfig 250J includes CodedVVecLength syntactic element 256, be set to indicate that in addition to element 1 to MinNumOfCoeffsForAmbHOA and Those of specified element, all members of V vector in ContAddAmbHoaChan syntactic element (in this example it is assumed that zero) Element is through decoding.The part HOAconfig 250J also includes the set of SpatialInterpolationMethod syntactic element 255 to refer to The interpolating function for showing space-time interpolation is raised cosine.In addition, the part HOAconfig 250J includes CodedSpatialInterpolationTime 254 is set to indicate that the interpolated sample duration for 256. The part HOAconfig 250J further comprises MinAmbHoaOrder syntactic element 150, is set to indicate that in environment HOA The MinimumHOA order of appearance is one, and wherein audio decoding apparatus 24 can export MinNumofCoeffsForAmbHOA grammer member Element is equal to (1+1)2Or four.The part HOAconfig 250J includes HoaOrder syntactic element 152, is set to indicate that content HOA order be equal to three (or in other words N=3), wherein audio decoding apparatus 24 can export NumOfHoaCoeffs equal to (N +1)2Or 16.

As further shown in the example of Figure 10 J (i), part 248J includes USAC-3D audio frame, two of them HOA frame 249E and 249F is stored in the load of USAC extremely efficient, it is assumed that when enabling frequency spectrum tape copy (SBR), described two audio frames It is stored in a USAC-3D frame.Audio decoding apparatus 24 can be exported as numHOATransportChannels syntactic element And several flexible transmission channels of the function of MinNumOfCoeffsForAmbHOA syntactic element.In the following example, it is assumed that NumHOATransportChannels syntactic element is equal to 7 and MinNumOfCoeffsForAmbHOA syntactic element and is equal to four, Wherein the number in flexible transmission channel is equal to numHOATransportChannels syntactic element and subtracts MinNumOfCoeffsForAmbHOA syntactic element (or three).

Frame 249E and 249F is described in more detail in Figure 10 J (ii).As shown in the example of Figure 10 J (ii), frame 249E includes CSID field 154 arrives 154C and VVectorData field 156 and 156B.CSID field 154 includes unitC 267, bb 266 And ba 265 is together with ChannelType 269, each of these person be set to respective value 01 shown in the example of Figure 10 J (i), 1,0 and 01.CSID field 154B includes unitC 267, bb 266 and ba 265 together with ChannelType 269, therein every One is set to respective value 01,1,0 and 01 shown in the example of Figure 10 J (ii).CSID field 154C includes with value 3 ChannelType field 269.CSID field 154 is corresponding corresponding to one in transmission channel 1,2 and 3 to each of 154C Person.

In the example of Figure 10 J (ii), frame 249E includes two signals based on vector (it is assumed that ChannelType 269 It is equal to 1) in CSID field 154 and 154B and a spacing wave is (it is assumed that ChannelType 269 is equal in CSID field 154C 3).The aforementioned part HOAconfig 250H is given, audio decoding apparatus 24 can determine that all 12 V vector elements are encoded (wherein 12 be to export as (HOAOrder+1)2(MinNumOfCoeffsForAmbHOA)-(ContAddAmbHoaChan)=16-4-0 =12).Therefore, VVectorData 156 and 156B respectively contains all 12 vector elements, and each of these person passes through 8 Position uniform quantization.As pointed by footnote 1, the number of VVectorData element and index are decoded by parameter CodedVVecLength=0 is specified.In addition, such as pointed by single asterisk (*), decoding scheme by NbitsQ=5 for pair It answers and is signaled in the CSID field of transmission channel.

In frame 249F, CSID field 154 and 154B in frame 249E CSID field 154 and 154B it is identical, and frame The CSID field 154C of 249F is switched to the ChannelType for one.Therefore, the CSID field 154C of frame 249B includes Cbflag 267, Pflag 267 (instruction huffman coding) and Nbits 261 (being equal to 12).As a result, frame 249F includes third VVectorData field 156C, it includes 12 V vector elements, and each of these person is by 12 position uniform quantizations and through suddenly Fu Man decoding.As noted above, through decoding the number of VVectorData element and index by parameter CodedVVecLength =0 is specified, and Hoffman decodeng scheme is led to by NbitsQ=12, CbFlag=0 and Pflag=0 for this specific transmission It is signaled in the CSID field 154C in road (for example, the 3rd transmission channel).

The example of Figure 10 K (i) and 10K (ii) illustrate to have generated the second example bit stream 248K and the adjoint portion HOA config Divide 250K with corresponding with the above situation 1 in the above pseudo-code.In the example of Figure 10 K (i), the part HOAconfig 250K packet Syntactic element containing CodedVVecLength 256 is set to indicate that in addition to element 1 to MinNumOfCoeffsForAmbHOA And those of specified element in ContAddAmbHoaChan syntactic element (in this example it is assumed that one), V vector own Element is through decoding.The part HOAconfig 250K also includes SpatialInterpolationMethod syntactic element 255, warp The interpolating function for being set to indicate that space-time interpolation is raised cosine.In addition, the part HOAconfig 250K includes CodedSpatialInterpolationTime 254 is set to indicate that the interpolated sample duration for 256.

The part HOAconfig 250K further comprises MinAmbHoaOrder syntactic element 150, is set to indicate that ring The MinimumHOA order of border HOA content is one, and wherein audio decoding apparatus 24 can export MinNumofCoeffsForAmbHOA Syntactic element is equal to (1+1)2Or four.Audio decoding apparatus 24 can also export MaxNoOfAddActiveAmbCoeffs grammer member Element, the difference being such as set between NumOfHoaCoeff syntactic element and MinNumOfCoeffsForAmbHOA, in this example It is assumed that it is equal to 16-4 or 12.Audio decoding apparatus 24 can also export AmbAsignmBits syntactic element, such as be set to ceil (log2 (MaxNoOfAddActiveAmbCoeffs))=ceil (log2 (12))=4.The part HOAconfig 250K includes HoaOrder syntactic element 152 is set to indicate that the HOA order of content is equal to three (or in other words N=3), middle pitch Frequency decoding apparatus 24 can export NumOfHoaCoeffs equal to (N+1)2Or 16.

As further shown in the example of Figure 10 K (i), part 248K includes USAC-3D audio frame, two of them HOA frame 249G and 249H is stored in the load of USAC extremely efficient, it is assumed that when enabling frequency spectrum tape copy (SBR), described two audio frames It is stored in a USAC-3D frame.Audio decoding apparatus 24 can be exported as numHOATransportChannels syntactic element And several flexible transmission channels of the function of MinNumOfCoeffsForAmbHOA syntactic element.In the following example, it is assumed that NumHOATransportChannels syntactic element is equal to 7 and MinNumOfCoeffsForAmbHOA syntactic element and is equal to four, Wherein the number in flexible transmission channel is equal to numHOATransportChannels syntactic element and subtracts MinNumOfCoeffsForAmbHOA syntactic element (or three).

Frame 249G and 249H is described in more detail in Figure 10 K (ii).As shown in the example of Figure 10 K (ii), frame 249G includes CSID field 154 arrives 154C and VVectorData field 156.CSID field 154 include CodedAmbCoeffIdx 246, (wherein double asterisk (* *) instruction is for the 1st flexible transmission channel, it is herein assumed that decoding by AmbCoeffIdxTransition 247 The internal state of device is AmbCoeffIdxTransitionState=2, causes CodedAmbCoeffIdx bit field in place Signal in stream or specify in other ways) and ChannelType 269 (its be equal to two, to signal correspondence Payload is additional environment HOA coefficient).In this example, AmbCoeffIdx can be exported as and is equal to by audio decoding apparatus 24 CodedAmbCoeffIdx+1+MinNumOfCoeffsForAmbHOA or 5.CSID field 154B includes unitC 267, bb Together with ChannelType 269, each of these person is set to corresponding shown in the example of Figure 10 K (ii) 266 and ba 265 Value 01,1,0 and 01.CSID field 154C includes the ChannelType field 269 with value 3.

In the example of Figure 10 K (ii), frame 249G includes the single signal based on vector (it is assumed that ChannelType 269 1) and a spacing wave (it is assumed that ChannelType 269 in CSID field 154C be equal to 3) it is equal in CSID field 154B.It gives The fixed aforementioned part HOAconfig 250K, audio decoding apparatus 24 can determine that 11 V vector elements are encoded and (wherein 12 export as (HOAOrder+1)2(MinNumOfCoeffsForAmbHOA)-(ContAddAmbHoaChan)=16-4-1=11).Cause This, VVectorData 156 includes all 11 vector elements, and each of these person passes through 8 position uniform quantizations.Such as footnote 1 It is pointed, it is decoded the number of VVectorData element and index and is specified by parameter CodedVVecLength=0.In addition, such as As pointed by footnote 2, decoding scheme is signaled in the CSID field for corresponding to transmission channel by NbitsQ=5.

In frame 249H, CSID field 154 includes the AmbCoeffIdxTransition that instruction has occurred without transformation 247, and therefore can from former frame imply without signal or in other ways again specify CodedAmbCoeffIdx 246.The CSID field 154B and 154C of frame 249H is identical as the CSID field of frame 249G, and thus, such as 249G at same frame, frame 249H includes single VVectorData field 156, and it includes 11 vector elements, each of these person passes through 8 position even amounts Change.

Figure 10 L (i) and 10L (ii) illustrates to have generated the first example bit stream 248L and adjoint HOA config part 250L With corresponding with the situation 2 in the above pseudo-code.In the example of Figure 10 L (i), the part HOAconfig 250L includes CodedVVecLength syntactic element 256 is set to indicate that in addition to 0 rank until by MinAmbHoaOrder syntactic element 150 (it is equal to (HoaOrder+1) in this example2-(MinAmbHoaOrder+1)2=16-4=12) specified order member Element, all elements of V vector are through decoding.The part HOAconfig 250L also includes SpatialInterpolationMethod language Method element 255 is set to indicate that the interpolating function of space-time interpolation is raised cosine.In addition, the part HOAconfig 250L includes CodedSpatialInterpolationTime 254, is set to indicate that and continues for 256 interpolated sample Time.The part HOAconfig 250L further comprises MinAmbHoaOrder syntactic element 150, is set to indicate that environment The MinimumHOA order of HOA content is one, and wherein audio decoding apparatus 24 can export MinNumofCoeffsForAmbHOA language Method element is equal to (1+1)2Or four.The part HOAconfig 250L includes HoaOrder syntactic element 152, is set to indicate that The HOA order of content is equal to three (or in other words N=3), and wherein audio decoding apparatus 24 can export NumOfHoaCoeffs etc. In (N+1)2Or 16.

As further shown in the example of Figure 10 L (i), part 248L includes USAC-3D audio frame, two of them HOA frame 249I and 249J is stored in the load of USAC extremely efficient, it is assumed that when enabling frequency spectrum tape copy (SBR), described two audio frames It is stored in a USAC-3D frame.Audio decoding apparatus 24 can be exported as numHOATransportChannels syntactic element And several flexible transmission channels of the function of MinNumOfCoeffsForAmbHOA syntactic element.In following instance, it is assumed that NumHOATransportChannels syntactic element is equal to 7 and MinNumOfCoeffsForAmbHOA syntactic element and is equal to four, Wherein the number in flexible transmission channel is equal to numHOATransportChannels syntactic element and subtracts MinNumOfCoeffsForAmbHOA syntactic element (or three).

Frame 249I and 249J is described in more detail in Figure 10 L (ii).As shown in the example of Figure 10 L (ii), frame 249I includes CSID field 154 arrives 154C and VVectorData field 156 and 156B.CSID field 154 includes unitC 267, bb 266 And ba 265 is together with ChannelType 269, each of these person be set to respective value 01 shown in the example of Figure 10 J (i), 1,0 and 01.CSID field 154B includes unitC 267, bb 266 and ba 265 together with ChannelType 269, therein every One is set to respective value 01,1,0 and 01 shown in the example of Figure 10 L (ii).CSID field 154C includes with value 3 ChannelType field 269.CSID field 154 is corresponding corresponding to one in transmission channel 1,2 and 3 to each of 154C Person.

In the example of Figure 10 L (ii), frame 249I includes two signals based on vector (it is assumed that ChannelType 269 It is equal to 1) in CSID field 154 and 154B and a spacing wave is (it is assumed that ChannelType 269 is equal in CSID field 154C 3).The aforementioned part HOAconfig 250H is given, audio decoding apparatus 24 can determine that 12 V vector elements are encoded.Therefore, VVectorData 156 and 156B respectively contains 12 vector elements, and each of these person passes through 8 position uniform quantizations.Such as foot Pointed by note 1, is decoded the number of VVectorData element and index and specified by parameter CodedVVecLength=0.In addition, Such as pointed by single asterisk (*), decoding scheme uses signal in the CSID field for corresponding to transmission channel by NbitsQ=5 Notice.

In frame 249J, CSID field 154 and 154B in frame 249I CSID field 154 and 154B it is identical, and frame The CSID field 154C of 249F is switched to the ChannelType for one.Therefore, the CSID field 154C of frame 249B includes Cbflag 267, Pflag 267 (instruction huffman coding) and Nbits 261 (being equal to 12).As a result, frame 249F includes third VVectorData field 156C, it includes 12 V vector elements, and each of these person is by 12 position uniform quantizations and through suddenly Fu Man decoding.As noted above, through decoding the number of VVectorData element and index by parameter CodedVVecLength =0 is specified, and Hoffman decodeng scheme is led to by NbitsQ=12, CbFlag=0 and Pflag=0 for this specific transmission It is signaled in the CSID field 154C in road (for example, the 3rd transmission channel).

The example of Figure 10 M (i) and 10M (ii) illustrate to have generated the second example bit stream 248M and the adjoint portion HOA config Divide 250M with corresponding with the above situation 2 in the above pseudo-code.In the example of Figure 10 M (i), the part HOAconfig 250M packet Syntactic element containing CodedVVecLength 256 is set to indicate that in addition to 0 rank until by MinAmbHoaOrder grammer member (it is equal to (HoaOrder+1) to element 150 in this example2-(MinAmbHoaOrder+1)2=16-4=12) specified order Element, all elements of V vector are through decoding.The part HOAconfig 250M also includes SpatialInterpolationMethod Syntactic element 255 is gathered to indicate the interpolating function of space-time interpolation as raised cosine.In addition, the part HOAconfig 250M packet Containing CodedSpatialInterpolationTime 254, it is set to indicate that the interpolated sample duration for 256.

The part HOAconfig 250M further comprises MinAmbHoaOrder syntactic element 150, is set to indicate that ring The MinimumHOA order of border HOA content is one, and wherein audio decoding apparatus 24 can export MinNumofCoeffsForAmbHOA Syntactic element is equal to (1+1)2Or four.Audio decoding apparatus 24 can also export MaxNoOfAddActiveAmbCoeffs grammer member Element, the difference being such as set between NumOfHoaCoeff syntactic element and MinNumOfCoeffsForAmbHOA, in this example It is assumed that it is equal to 16-4 or 12.Audio decoding apparatus 24 can also export AmbAsignmBits syntactic element, such as be set to ceil (log2 (MaxNoOfAddActiveAmbCoeffs))=ceil (log2 (12))=4.The part HOAconfig 250M includes HoaOrder syntactic element 152 is set to indicate that the HOA order of content is equal to three (or in other words N=3), middle pitch Frequency decoding apparatus 24 can export NumOfHoaCoeffs equal to (N+1)2Or 16.

As further shown in the example of Figure 10 M (i), part 248M includes USAC-3D audio frame, two of them HOA frame 249K and 249L is stored in the load of USAC extremely efficient, it is assumed that when enabling frequency spectrum tape copy (SBR), described two audio frames It is stored in a USAC-3D frame.Audio decoding apparatus 24 can be exported as numHOATransportChannels syntactic element And several flexible transmission channels of the function of MinNumOfCoeffsForAmbHOA syntactic element.In the following example, it is assumed that NumHOATransportChannels syntactic element is equal to 7 and MinNumOfCoeffsForAmbHOA syntactic element and is equal to four, Wherein the number in flexible transmission channel is equal to numHOATransportChannels syntactic element and subtracts MinNumOfCoeffsForAmbHOA syntactic element (or three).

Frame 249K and 249L is described in more detail in Figure 10 M (ii).As shown in the example of Figure 10 M (ii), frame 249K includes CSID field 154 arrives 154C and VVectorData field 156.CSID field 154 include CodedAmbCoeffIdx 246, (wherein double asterisk (* *) instruction is for the 1st flexible transmission channel, it is herein assumed that decoding by AmbCoeffIdxTransition 247 The internal state of device is AmbCoeffIdxTransitionState=2, causes CodedAmbCoeffIdx bit field in place Signal in stream or specify in other ways) and ChannelType 269 (its be equal to two, to signal correspondence Payload is additional environment HOA coefficient).In this example, AmbCoeffIdx can be exported as and is equal to by audio decoding apparatus 24 CodedAmbCoeffIdx+1+MinNumOfCoeffsForAmbHOA or 5.CSID field 154B includes unitC 267, bb Together with ChannelType 269, each of these person is set to corresponding shown in the example of Figure 10 M (ii) 266 and ba 265 Value 01,1,0 and 01.CSID field 154C includes the ChannelType field 269 with value 3.

In the example of Figure 10 M (ii), frame 249K includes the single signal based on vector (it is assumed that ChannelType 269 1) and a spacing wave (it is assumed that ChannelType 269 in CSID field 154C be equal to 3) it is equal in CSID field 154B.It gives The fixed aforementioned part HOAconfig 250M, audio decoding apparatus 24 can determine that 12 V vector elements are encoded.Therefore, VVectorData 156 includes 12 vector elements, and each of these person passes through 8 position uniform quantizations.As pointed by footnote 1, It is decoded the number of VVectorData element and index and is specified by parameter CodedVVecLength=0.In addition, such as by footnote 2 Pointed, decoding scheme is signaled in the CSID field for corresponding to transmission channel by NbitsQ=5.

In frame 249L, CSID field 154 includes the AmbCoeffIdxTransition that instruction has occurred without transformation 247, and therefore can from former frame imply without signal or in other ways again specify CodedAmbCoeffIdx 246.The CSID field 154B and 154C of frame 249L is identical as the CSID field of frame 249K, and thus, such as 249K at same frame, frame 249L includes single VVectorData field 156, and it includes all 12 vector elements, each of these person is equal by 8 positions Even quantization.

Figure 10 N (i) and 10N (ii) illustrates to have generated the first example bit stream 248N and adjoint HOA config part 250N With corresponding with the situation 3 in the above pseudo-code.In the example of Figure 10 N (i), the part HOAconfig 250N includes CodedVVecLength syntactic element 256, be set to indicate that in addition to ContAddAmbHoaChan syntactic element (its Those of specified element in zero) is assumed in this example, all elements of V vector are through decoding.The part HOAconfig 250N includes The set of SpatialInterpolationMethod syntactic element 255 is to indicate the interpolating function of space-time interpolation as more than rising String.In addition, the part HOAconfig 250N includes CodedSpatialInterpolationTime 254, it is set to indicate that For 256 interpolated sample duration.The part HOAconfig 250N further comprises MinAmbHoaOrder syntactic element 150, it is set to indicate that the MinimumHOA order of environment HOA content is one, wherein audio decoding apparatus 24 can export MinNumofCoeffsForAmbHOA syntactic element is equal to (1+1)2Or four.The part HOAconfig 250N includes HoaOrder language Method element 152 is set to indicate that the HOA order of content is equal to three (or in other words N=3), wherein audio decoding apparatus 24 can export NumOfHoaCoeffs equal to (N+1)2Or 16.

As further shown in the example of Figure 10 N (i), part 248N includes USAC-3D audio frame, two of them HOA frame 249M and 249N is stored in the load of USAC extremely efficient, it is assumed that when enabling frequency spectrum tape copy (SBR), described two audio frames It is stored in a USAC-3D frame.Audio decoding apparatus 24 can be exported as numHOATransportChannels syntactic element And several flexible transmission channels of the function of MinNumOfCoeffsForAmbHOA syntactic element.In following instance, it is assumed that NumHOATransportChannels syntactic element is equal to 7 and MinNumOfCoeffsForAmbHOA syntactic element and is equal to four, Wherein the number in flexible transmission channel is equal to numHOATransportChannels syntactic element and subtracts MinNumOfCoeffsForAmbHOA syntactic element (or three).

Frame 249M and 249N is described in more detail in Figure 10 N (ii).As shown in the example of Figure 10 N (ii), frame 249M includes CSID field 154 arrives 154C and VVectorData field 156 and 156B.CSID field 154 includes unitC 267, bb 266 And ba 265 is together with ChannelType 269, each of these person be set to respective value 01 shown in the example of Figure 10 J (i), 1,0 and 01.CSID field 154B includes unitC 267, bb 266 and ba 265 together with ChannelType 269, therein every One is set to respective value 01,1,0 and 01 shown in the example of Figure 10 N (ii).CSID field 154C includes with value 3 ChannelType field 269.CSID field 154 is corresponding corresponding to one in transmission channel 1,2 and 3 to each of 154C Person.

In the example of Figure 10 N (ii), frame 249M includes two signals based on vector (it is assumed that ChannelType 269 It is equal to 1) in CSID field 154 and 154B and a spacing wave is (it is assumed that ChannelType 269 is equal in CSID field 154C 3).The aforementioned part HOAconfig 250M is given, audio decoding apparatus 24 can determine that 16 V vector elements are encoded.Therefore, VVectorData 156 and 156B respectively contains 16 vector elements, and each of these person passes through 8 position uniform quantizations.Such as foot Pointed by note 1, is decoded the number of VVectorData element and index and specified by parameter CodedVVecLength=0.In addition, Such as pointed by single asterisk (*), decoding scheme uses signal in the CSID field for corresponding to transmission channel by NbitsQ=5 Notice.

In frame 249N, CSID field 154 and 154B in frame 249M CSID field 154 and 154B it is identical, and frame The CSID field 154C of 249F is switched to the ChannelType for one.Therefore, the CSID field 154C of frame 249B includes Cbflag 267, Pflag 267 (instruction huffman coding) and Nbits 261 (being equal to 12).As a result, frame 249F includes third VVectorData field 156C, it includes 16 V vector elements, and each of these person is by 12 position uniform quantizations and through suddenly Fu Man decoding.As noted above, through decoding the number of VVectorData element and index by parameter CodedVVecLength =0 is specified, and Hoffman decodeng scheme is led to by NbitsQ=12, CbFlag=0 and Pflag=0 for this specific transmission It is signaled in the CSID field 154C in road (for example, the 3rd transmission channel).

The example of Figure 10 O (i) and 10O (ii) illustrate to have generated the second example bit stream 248O and the adjoint portion HOA config Divide 250O with corresponding with the above situation 3 in the above pseudo-code.In the example of Figure 10 O (i), the part HOAconfig 250O packet Syntactic element containing CodedVVecLength 256 is set to indicate that in addition in ContAddAmbHoaChan syntactic element (its In this example it is assumed that in one) specify those of element, all elements of V vector are through decoding.The part HOAconfig 250O is also Gather comprising SpatialInterpolationMethod syntactic element 255 with the interpolating function for indicating space-time interpolation and is Raised cosine.In addition, the part HOAconfig 250O includes CodedSpatialInterpolationTime 254, it is set to It is designated as 256 interpolated sample duration.

The part HOAconfig 250O further comprises MinAmbHoaOrder syntactic element 150, is set to indicate that ring The MinimumHOA order of border HOA content is one, and wherein audio decoding apparatus 24 can export MinNumofCoeffsForAmbHOA Syntactic element is equal to (1+1)2Or four.Audio decoding apparatus 24 can also export MaxNoOfAddActiveAmbCoeffs grammer member Element, the difference being such as set between NumOfHoaCoeff syntactic element and MinNumOfCoeffsForAmbHOA, in this example It is assumed that it is equal to 16-4 or 12.Audio decoding apparatus 24 can also export AmbAsignmBits syntactic element, such as be set to ceil (log2 (MaxNoOfAddActiveAmbCoeffs))=ceil (log2 (12))=4.The part HOAconfig 250O includes HoaOrder syntactic element 152 is set to indicate that the HOA order of content is equal to three (or in other words N=3), middle pitch Frequency decoding apparatus 24 can export NumOfHoaCoeffs equal to (N+1)2Or 16.

As further shown in the example of Figure 10 O (i), part 248O includes USAC-3D audio frame, two of them HOA frame 249O and 249P is stored in the load of USAC extremely efficient, it is assumed that when enabling frequency spectrum tape copy (SBR), described two audio frames It is stored in a USAC-3D frame.Audio decoding apparatus 24 can be exported as numHOATransportChannels syntactic element And several flexible transmission channels of the function of MinNumOfCoeffsForAmbHOA syntactic element.In following instance, it is assumed that NumHOATransportChannels syntactic element is equal to 7 and MinNumOfCoeffsForAmbHOA syntactic element and is equal to four, Wherein the number in flexible transmission channel is equal to numHOATransportChannels syntactic element and subtracts MinNumOfCoeffsForAmbHOA syntactic element (or three).

Frame 249O and 249P is described in more detail in Figure 10 O (ii).As shown in the example of Figure 10 O (ii), frame 249O includes CSID field 154 arrives 154C and VVectorData field 156.CSID field 154 include CodedAmbCoeffIdx 246, (wherein double asterisk (* *) instruction is for the 1st flexible transmission channel, it is herein assumed that decoding by AmbCoeffIdxTransition 247 The internal state of device is AmbCoeffIdxTransitionState=2, causes CodedAmbCoeffIdx bit field in place Signal in stream or specify in other ways) and ChannelType 269 (its be equal to two, to signal correspondence Payload is additional environment HOA coefficient).In this example, AmbCoeffIdx can be exported as and is equal to by audio decoding apparatus 24 CodedAmbCoeffIdx+1+MinNumOfCoeffsForAmbHOA or 5.CSID field 154B includes unitC 267, bb Together with ChannelType 269, each of these person is set to corresponding shown in the example of Figure 10 O (ii) 266 and ba 265 Value 01,1,0 and 01.CSID field 154C includes the ChannelType field 269 with value 3.

In the example of Figure 10 O (ii), frame 249O includes the single signal based on vector (it is assumed that ChannelType 269 1) and a spacing wave (it is assumed that ChannelType 269 in CSID field 154C be equal to 3) it is equal in CSID field 154B.It gives The fixed aforementioned part HOAconfig 250O, audio decoding apparatus 24 can determine that 16 subtract by ContAddAmbHoaChan syntactic element (for example, being wherein designated as ContAddAmbHoaChan syntactic element with 6 associated vector elements of index) specified one, Or 15 V vector elements are encoded.Therefore, VVectorData 156 includes 15 vector elements, and each of these person passes through 8 A uniform quantization.As pointed by footnote 1, the number of VVectorData element and index are decoded by parameter CodedVVecLength=0 is specified.In addition, decoding scheme is passed by NbitsQ=5 for corresponding as pointed by footnote 2 It is signaled in the CSID field in defeated channel.

In frame 249P, CSID field 154 includes the AmbCoeffIdxTransition that instruction has occurred without transformation 247, and therefore can from former frame imply without signal or in other ways again specify CodedAmbCoeffIdx 246.The CSID field 154B and 154C of frame 249P is identical as the CSID field of frame 249O, and thus, such as 249O at same frame, frame 249P includes single VVectorData field 156, and it includes 15 vector elements, each of these person passes through 8 position even amounts Change.

Figure 11 A to 11G is the frame that the various units of audio decoding apparatus 24 shown in the example of Fig. 5 are described in more detail Figure.Figure 11 A is the block diagram that the extraction unit 72 of audio decoding apparatus 24 is described in more detail.As shown in the example of Figure 11 A, extract Unit 72 may include that mode dissects unit 270, mode configuration unit 272 (" mode configuration unit 272 ") and configurable extraction Unit 274.

Mode, which dissects unit 270, can indicate to be configured to dissect instruction to be encoded HOA coefficient 11 so as to form The syntactic element of the decoding mode of bit stream 21 being indicated above is (for example, ChannelType grammer shown in the example of Figure 10 E Element) unit.Mode, which dissects unit 270, can will determine that syntactic element is transmitted to mode configuration unit 272.Mode configuration unit 272 can indicate to be configured to the unit based on configurable extraction unit 274 is configured through dissecting syntactic element.Mode configuration unit 272 configurable extraction units 274 are to extract being indicated or be based on through decoding based on direction for HOA coefficient 11 from bit stream 21 It is dissected syntactic element and extracts indicating based on vector through decoding for HOA coefficient 11 from bit stream 21.

When executing the coding based on direction, configurable extraction unit 274 can extract the version based on direction of HOA coefficient 11 (it is expressed as the information based on direction in the example of Figure 11 A for this and syntactic element associated with this encoded version 91).This information 91 based on direction may include in the example of directivity information 253 and Figure 10 E shown in the example of Figure 10 D Shown in the SideChannelInfoData based on direction, such as defined by null ChannelType.

When syntactic element instruction HOA coefficient 11 is encoded using the synthesis based on vector (for example, working as ChannelType syntactic element is equal to for the moment), it is extractable through decoding prospect V [k] vector 57, warp knit to can configure extraction unit 274 Code surrounding HOA coefficient 59 and encoded nFG signal 59.Configurable extraction unit 274 can also indicate HOA determining syntactic element Coefficient 11 is extracted at once from bit stream 21 later using the synthesis based on vector is encoded CodedSpatialInterpolationTime syntactic element 254 and SpatialInterpolationMethod syntactic element 255, these syntactic elements 254 and 255 are transmitted to space-time interpolation unit 76.

Figure 11 B is the block diagram that the quantifying unit 74 of audio decoding apparatus 24 shown in the example of Fig. 5 is described in more detail.Amount Changing unit 74 can indicate to be configured and operate in the reciprocal mode of quantifying unit 52 shown in the example with Fig. 4 so as to through translating Code prospect V [k] vector 57 carries out entropy decoding and de-quantization and is generated in turn through reducing prospect V [k] vector 55kUnit.Scalar/ Entropy dequantizing unit 984 may include classification/residual decoding unit 276, predicting unit 278 and uniform dequantizing unit 280.

Classification/residual decoding unit 276 can indicate to be configured to using (as noted above by huffman table information 241 Out, the syntactic element being expressed as in bit stream 21) identification huffman table relative to through decoding prospect V [k] vector 57 execute suddenly The graceful decoded unit of husband.Quantified prospect V [k] vector can be output to predicting unit 278 by classification/residual decoding unit 276.In advance Surveying unit 278 can indicate to be configured to execute prediction relative to quantified prospect V [k] vector thus will based on prediction mode 237 Enhanced quantified prospect V [k] vector is output to the unit of uniform dequantizing unit 280.Uniform dequantizing unit 280 can indicate It is configured to execute de-quantization relative to enhanced quantified prospect V [k] vector based on nbits value 233 to which output is through reducing Prospect V [k] vector 55kUnit.

Figure 11 C is the psychologic acoustics decoding unit 80 that audio decoding apparatus 24 shown in the example of Fig. 5 is described in more detail Block diagram.As noted above, psychologic acoustics decoding unit 80 can psychologic acoustics audio coding shown in the example with Fig. 4 The reciprocal mode of unit 40 operates to be decoded encoded surrounding HOA coefficient 59 and encoded nFG signal 61 and in turn Generating energy compensated surrounding HOA coefficient 47' and interpolated nFG signal 49'(, it is also known as interpolated nFG audio pair As 49').The compensated surrounding HOA coefficient 47' of energy can be transmitted to HOA coefficient and work out unit by psychologic acoustics decoding unit 80 82 and nFG signal 49' is transmitted to the device 84 that reorders.Psychologic acoustics decoding unit 80 may include being similar to psychologic acoustics audio to translate Multiple audio decoders 80 of code unit 40 arrive 80N.Audio decoder 80 can pass through psychologic acoustics audio to 80N with sufficient amount Decoding unit 40 executes individuation or is contained in psychologic acoustics audio coding unit 40 in other ways with as noted above It supports to decode while each signal to each channel of background HOA coefficient 47' and nFG signal 49'.

Figure 11 D is the block diagram that the unit 84 that reorders of audio decoding apparatus 24 shown in the example of Fig. 5 is described in more detail. The unit 84 that reorders can indicate to be configured and by with above in relation to substantially reciprocal side in a manner of reordering described in unit 34 The unit of formula operation.The unit 84 that reorders may include that vector reorders unit 282, can indicate to be configured to receive instruction HOA The unit of the syntactic element 205 of the original order of the prospect component of coefficient 11.Extraction unit 72 can dissect this from bit stream 21 A little syntactic elements 205, and syntactic element 205 is transmitted to the unit 84 that reorders.Vector reorders unit 282 can be heavy based on these Syntactic element 205 sort to interpolated nFG signal 49' and through reducing prospect V [k] vector 55kIt reorders to generate rearranged sequence " and prospect V [k] vector 55 of rearranged sequence of nFG signal 49k'.Reordering unit 84 can be by the nFG signal 49 of rearranged sequence " It is output to prospect and works out unit 78, and by prospect V [k] vector 55 of rearranged sequencek'It is output to space-time interpolation unit 76.

Figure 11 E is the space-time interpolation unit 76 that audio decoding apparatus 24 shown in the example of Fig. 5 is described in more detail Block diagram.Space-time interpolation unit 76 can be similar to above in relation to mode described by space-time interpolation unit 50 Mode operates.Space-time interpolation unit 76 may include V interpolation unit 284, can indicate to be configured to receive rearranged sequence Prospect V [k] vector 55k' and prospect V [k] vector 55 relative to rearranged sequencek' and rearranged sequence prospect V [k-1] vector 55k-1' space-time interpolation is executed to generate interpolated prospect V [k] vector 55k"Unit.V interpolation unit 284 can be based on CodedSpatialInterpolationTime syntactic element 254 and SpatialInterpolationMethod syntactic element 255 execute interpolation.In some instances, V interpolation unit 285 can be used by SpatialInterpolationMethod grammer member The type of interpolation of 255 identification of element is in the duration specified by CodedSpatialInterpolationTime syntactic element 254 Interpolation V vector.Space-time interpolation unit 76 can be by interpolated prospect V [k] vector 55k"It is forwarded to prospect and works out unit 78.

Figure 11 F is the frame that the prospect formulation unit 78 of audio decoding apparatus 24 shown in the example of Fig. 5 is described in more detail Figure.It may include multiplication unit 286 that prospect, which works out unit 78, can indicate to be configured to relative to interpolated prospect V [k] vector 55k"And rearranged sequence nFG signal 49 " executes matrix multiplication to generate the unit of prospect HOA coefficient 65.

Figure 11 G is the HOA coefficient formulation unit 82 that audio decoding apparatus 24 shown in the example of Fig. 5 is described in more detail Block diagram.It may include addition unit 288 that HOA coefficient, which works out unit 82, can indicate to be configured to for prospect HOA coefficient 65 being added to Surrounding HOA channel 47' is to obtain the unit of HOA coefficient 11'.

Figure 12 is the figure of the example audio ecosystem for the various aspects that technology described in the present invention can be performed in explanation.Such as Shown in Figure 12, the audio ecosystem 300 may include acquisition 301, editor 302, decoding 303, transmitting 304, and reset 305.

The technology of acquisition audio content of the audio ecosystem 300 can be indicated by obtaining 301.Obtain 301 example include (but It is not limited to) sound (for example, live sound), audio generation are recorded (for example, the production of audio object, musical drama, sound rendering, mould It is quasi-), etc..In some instances, sound can be in concert, sport event and the record when exercising supervision.In some instances, Can execute simulation when generate and production/mixing (for example, film, game) audio.Audio object can be used for Hollywood (for example, The operating room IMAX) in.In some instances, obtaining 301 can be executed by creator of content (such as creator of content 12 of Fig. 3).

Editor 302 can indicate the editor of the audio ecosystem 300 and/or modify the technology of audio content.As a reality Example, can edit audio content and multiple unit groups of audio content are combined into the individual unit of audio content.As another Example can edit by adjusting actual audio content (for example, level of one or more frequency components of adjustment audio content) Audio content.In some instances, editor 302 can be executed by audio editing system (such as audio editing system 18 of Fig. 3).? In some examples, can in mobile device (such as one or more of mobile device illustrated in Figure 29) executive editor 302.

Decoding 303 can indicate the technology of the expression that audio content is decoded as to audio content of the audio ecosystem 300.? In some examples, the expression of audio content can be bit stream, such as the bit stream 21 of Fig. 3.In some instances, decoding 302 can be by example 20 audio coding apparatus of audio coding apparatus of such as Fig. 3 executes.

What transmitting 304 can indicate the audio ecosystem 300 is transmitted to content consumer from creator of content for audio content Element.In some instances, real-time or near real-time audio content can be conveyed.For example, audio content can crossfire be transferred to Content consumer.In some instances, audio content can be by being decoded in media (such as computer-readable storage for audio content Media) on and conveyed.For example, audio content is storable in CD, driver and the like (for example, Blu-ray Disc, storage Device card, hard disk drive etc.) on.

Reproducing audio content and resetting to the technology of content consumer for the audio ecosystem 300 can be indicated by resetting 305. In some instances, resetting 305 may include that the one or more aspects based on playback environment reproduce 3D sound field.In other words, it resets 305 can based on local acoustical scape.

Figure 13 is the figure that an example of the audio ecosystem of Figure 12 is described in more detail.As illustrated in Figure 13, audio The ecosystem 300 may include audio content 308, film workshop 310, music studio 311, gaming audio operating room 312, base Audio content 313, decoding engine 314, gaming audio primary sound 315, gaming audio decoding/reproduction engine 316 in channel, and pass Send system 317.Illustrated example gaming audio operating room 312 in Figure 26.Illustrate some example gaming audio decoding/reproductions in Figure 27 Engine 316.

As illustrated in fig. 13, film workshop 310, music studio 311 and gaming audio operating room 312 can receive audio Content 308.In a certain example, audio content 308 can indicate the output of the acquisition 301 of Figure 12.Film workshop 310 can be such as Audio content 313 (for example, in 2.0,5.1 and 7.1) based on channel is exported by using Digital Audio Workstation (DAW). Music studio 310 can for example export the audio content 313 (for example, in 2.0 and 5.1) based on channel by using DAW. In any case, decoding engine 314 can receive the audio content 313 based on channel and be based on one or more codec (examples Such as, the true HD of AAC, AC3, Doby, Dolby Digital be subject to and DTS main audio) audio content 313 based on channel is encoded For being exported by delivery system 317.By this method, decoding engine 314 can be the example of the decoding 303 of Figure 12.Gaming audio Operating room 312 for example can export one or more gaming audio primary sounds 315 by using DAW.Gaming audio decoding/reproduction engine Audio primary sound 315 can be decoded and/or be reproduced as the audio content based on channel to be used to be exported by delivery system 317 by 316.? In some examples, film workshop 310, music studio 311 and the output of gaming audio operating room 312 can indicate the volume of Figure 12 Collect 302 output.In some instances, decoding engine 314 and/or the output of gaming audio decoding/reproduction engine 316 can be via The technology of the transmitting 304 of Figure 12 is transported to delivery system 317.

Figure 14 is the figure that another example of the audio ecosystem of Figure 12 is described in more detail.As illustrated in Figure 14, audio Ecosystem 300B may include broadcast recoding audio object 319, professional audio systems 320, capture 322, HOA on consumer devices 324, consumer audio, TV and attachment 325 and automobile audio system 326 are reproduced on audio format 323, device.

As illustrated in Figure 14, it is captured on broadcast recoding audio object 319, professional audio systems 320 and consumer devices 322 can all export it using HOA audio format 323 and decode.By this method, HOA audio lattice can be used in audio content Formula 323, which is decoded as can be used, reproduces 324, consumer audio, TV and attachment 325 and the playback of automobile audio system 326 on device Single expression.In other words, the single expression of audio content can be reset at universal audio playback system (that is, with example is required Such as 5.1,7.1 specific configurations are opposite).

Figure 15 A and 15B are the figures that other examples of the audio ecosystem of Figure 12 are described in more detail.As described in Figure 15 A Bright, audio ecosystem 300C may include obtaining element 331 and playback element 336.Obtain element 331 may include it is wired and/or Surround sound capture 334 and mobile device 335 are (for example, intelligence on wireless acquisition device 332 (for example, intrinsic microphone), device Phone and tablet computer).In some instances, wiredly and/or wirelessly acquisition device 332 can be via wiredly and/or wirelessly leading to It is coupled to mobile device 335 in letter channel 333.

One or more technologies according to the present invention, mobile device 335 can be used to obtain sound field.For example, mobile device 335 can Via surround sound capture 334 on wiredly and/or wirelessly acquisition device 332 and/or device (for example, being integrated into mobile device 335 Multiple microphones) obtain sound field.Acquired sound field then can be decoded into HOA 337 for by resetting member by mobile device 335 One or more of part 336 is reset.For example, mobile device 335 user can record live events (for example, meet, meeting, play, Concert etc.) (obtaining its sound field), and the record is decoded into HOA.

Mobile device 335 can also reset HOA through decoding sound field using one or more of element 336 is reset.For example, moving Dynamic device 335 can be decoded HOA through decoding sound field, and will to reset one or more of element 336 reconstruction sound field Signal, which is output to, resets one or more of element 336.As an example, mobile device 335 can be using wirelessly and/or wirelessly Communication channel 338 outputs a signal to one or more loudspeakers (for example, loudspeaker array, sound stick (sound bar), etc.).Make For another example, mobile device 335 can output a signal to one or more docks and/or one using docking solution 339 Or the loudspeaker (for example, audio system in intelligent automobile and/or family) of multiple docking.As another example, mobile device 335, which can reproduce 340 using head-telephone, (for example) generates actual ears to output a signal to one group of head-telephone Sound.

In some instances, specific mobile device 335 can obtain 3D sound field and reset same 3D sound field in the time later. In some instances, mobile device 335 can obtain 3D sound field, 3D sound field is encoded to HOA, and encoded 3D sound field is emitted to One or more other devices (for example, other mobile devices and/or other nonmobile devices) are for resetting.

As illustrated in Figure 15 B, audio ecosystem 300D may include audio content 343, game studios 344, through translating Code audio content 345, reproduction engine 346 and delivery system 347.In some instances, game studios 344 may include that can prop up Hold one or more DAW to the editor of HOA signal.For example, one or more described DAW may include can be configured with one or Multiple gaming audio systems operate the HOA plug-in unit and/or tool of (for example, work) together.In some instances, game studios The 344 exportable new primary sound formats for supporting HOA.Under any circumstance, game studios 344 can will be through decoding audio content 345 It is output to reproduction engine 346, reproduction engine 346 reproduce sound field for being reset by delivery system 347.

Figure 16 is the figure of the example audio code device for the various aspects that technology described in the present invention can be performed in explanation.Such as Illustrated in fig. 16, audio ecosystem 300E may include original 3D audio content 351, encoder 352, bit stream 353, decoder 354, reconstructor 355 and playback element 356.If Figure 16 further illustrate, encoder 352 may include Analysis of The Acoustic Fields and decompose 357, Background extracting 358, background high-lighting determine 359, audio coding 360, prospect/different audio extraction 361 and audio coding 362. In some instances, encoder 352 can be configured to perform the operation of the audio coding apparatus 20 similar to Fig. 3 and 4.Some In example, Analysis of The Acoustic Fields can be configured to perform the operation of the Analysis of The Acoustic Fields unit 44 similar to Fig. 4 with decomposition 357.Some In example, background extracting 358 and background high-lighting determine that 359 can be configured to perform the BG selecting unit 48 similar to Fig. 4 Operation.In some instances, audio coding 360 and audio coding 362 can be configured to perform the psychologic acoustics sound similar to Fig. 4 The operation of frequency translator unit 40.In some instances, prospect/different audio extraction 361 can be configured to perform similar to Fig. 4 Foreground selection unit 36 operation.

In some instances, prospect/different audio extraction 361 can correspond to 390 analyzing audio content of video frame of Figure 33. For example, prospect/different audio extraction 361 can determine that the audio content corresponding to area 391A to 391C is prospect audio.

As illustrated in fig. 16, encoder 352 can be configured with by can have 25 to 75Mbps bit rate it is original in Hold 351 bit streams 353 for being encoded to the bit rate that there can be 256kbps to 1.2Mbps.Figure 17 is the audio that Figure 16 is described in more detail The figure of one example of code device.

Figure 18 is the figure of the example audio decoding device for the various aspects that technology described in the present invention can be performed in explanation.Such as Illustrated in fig. 18, audio ecosystem 300E may include original 3D audio content 351, encoder 352, bit stream 353, decoder 354, reconstructor 355 and playback element 356.As Figure 16 further illustrates that decoder 354 may include audio decoder 363, audio Decoder 364, prospect rebuild 365 and mixing 366.In some instances, decoder 354 can be configured to perform similar to Fig. 3 With the operation of 5 audio decoding apparatus 24.In some instances, audio decoder 363, audio decoder 364 can be configured with Perform similarly to the operation of the psychologic acoustics decoding unit 80 of Fig. 5.In some instances, prospect is rebuild 365 and be can be configured to hold Row works out the operation of unit 78 similar to the prospect of Fig. 5.

As illustrated in fig. 16, decoder 354 can be configured to receive and decode bit stream 353 and by the reconstructed 3D sound of gained Field is output to reconstructor 355, and the reconstructor 355 then can cause to reset in the original 3D of one or more of element 356 output Hold 351 expression.Figure 19 is the figure that an example of audio decoding apparatus of Figure 18 is described in more detail.

Figure 20 A to 20G is that the example audio for the various aspects that technology described in the present invention can be performed in explanation obtains dress The figure set.Figure 20 A illustrates to may include the intrinsic microphone 370 for being jointly configured to multiple microphones of record 3D sound field.? In some examples, the multiple microphone of intrinsic microphone 370 can be located at the substantial spherical balls of the radius with about 4cm Surface on.In some instances, audio coding apparatus 20 can be integrated into intrinsic microphone so as to directly from microphone 370 it is defeated Bit stream 17 out.

Figure 20 B explanation can be configured to receive letter from one or more microphones such as such as one or more intrinsic microphones 370 Number production vehicle 372.Making vehicle 372 also may include audio coder, such as the audio coder 20 of Fig. 3.

Figure 20 C to 20E illustrates to may include the mobile device for being jointly configured to multiple microphones of record 3D sound field 374.In other words, the multiple microphone can have X, Y, Z diversity.In some instances, mobile device 374 may include Mike Wind 376 can be rotated relative to one or more other microphones of mobile device 374 to provide X, Y, Z diversity.Mobile device 374 also may include audio coder, such as the audio coder 20 of Fig. 3.

Figure 20 F illustrates the reinforcement type video capture device 378 that may be configured to record 3D sound field.In some instances, add Curable type video capture device 378 could attach to the helmet for participating in movable user.For example, reinforcement type video capture device 378 can be attached to the helmet of user when user goes boating.By this method, reinforcement type video capture device 378 can capture expression and use The 3D sound field of movement (for example, shock of the water behind user, another person of going boating are spoken in front of user) around family.

Figure 20 G illustrates the enhanced mobile device 380 of the attachment that may be configured to record 3D sound field.In some instances, it moves Dynamic device 380 can be similar to the mobile device 335 of Figure 15, wherein adding one or more attachmentes.For example, intrinsic microphone can The mobile device 335 of Figure 15 is attached to form the enhanced mobile device 380 of attachment.By this method, the enhanced mobile dress of attachment The higher-quality 3D sound field of voice capture component that the enhanced mobile device 380 of attachment is integrated into than being used only can be captured by setting 380 Version.

Figure 21 A to 21E is that the example audio for the various aspects that technology described in the present invention can be performed in explanation resets dress The figure set.Figure 21 A and 21B illustrate multiple loudspeakers 382 and sound stick 384.One or more technologies according to the present invention, loudspeaker 382 and/or sound stick 384 may be disposed to any arbitrary disposition while still resetting 3D sound field.Figure 21 C to 21E illustrate it is multiple wear by It talks about device replay device 386 and arrives 386C.Head-telephone replay device 386 can be coupled to solution via wired or wireless connection to 386C Code device.One or more technologies according to the present invention, can using the single general-purpose of sound field indicate come loudspeaker 382, sound stick 384 and Reproduced sound-field on head-telephone replay device 386 to any combination of 386C.

Figure 22 A to 22H be illustrate according to the present invention described in one or more technologies example audio playback environment Figure.For example, Figure 22 A illustrates that 5.1 speaker playback environment, Figure 22 B illustrate 2.0 (for example, stereo) speaker playback rings Border, Figure 22 C illustrate that the 9.1 speaker playback environment with full-height front speakers, Figure 22 D and 22E respectively illustrate that 22.2 raise Sound device playback environment, Figure 22 F illustrate that 16.0 speaker playback environment, Figure 22 G illustrate auto loud hailer playback environment, and Figure 22 H Illustrate the mobile device with earplug playback environment.

One or more technologies according to the present invention can be indicated using the single general-purpose of sound field come described in Figure 22 A to 22H The upper reproduced sound-field of any one of bright playback environment.In addition, technology of the invention enable reconstructor from generic representation again Show sound field for resetting in the playback environment for the environment being different from illustrated in Figure 22 A to 22H.For example, if design Consider to forbid appropriate storing of the loudspeaker according to 7.1 speaker playback environment (for example, if right surround loudspeaking can not be put Device), then technology of the invention enables reconstructor to compensate with other 6 loudspeakers, so that can think highly of in 6.1 loudspeakings It puts and environmentally realizes playback.

As illustrated in fig. 23, user can watch sports tournament when wearing head-telephone 386.According to the present invention one Or multiple technologies, the 3D sound field of sports tournament can be obtained (for example, one or more intrinsic microphones can be placed on the stick illustrated in Figure 24 In ball stadium and/or around), it can get the HOA coefficient corresponding to 3D sound field and be transferred to decoder, decoder can be based on HOA coefficient, which determines, to be rebuild 3D sound field and reconstructed 3D sound field is output to reconstructor, and reconstructor can get about playback environment The instruction of type (for example, head-telephone), and reconstructed 3D sound field is reproduced as head-telephone is caused to export sports tournament 3D sound field expression signal.In some instances, reconstructor can according to fig. 25 technology obtain class about playback environment The instruction of type.By this method, reconstructor can be for various loudspeaker positions, number, type, size and " adaptation ", and also manages Think for local environment and it is balanced.

Figure 28 be illustrate according to the present invention described in one or more technologies can by head-telephone simulate loudspeaking The figure of device configuration.As illustrated by Figure 28, technology of the invention aloows the user for wearing head-telephone 389 to experience sound , as the sound field is as being reset loudspeaker 388.By this method, user can listen to 3D sound field without sound is defeated Big region is arrived out.

Figure 30 be explanation with can according to the present invention described in the 3D sound field that is pocessed of one or more technologies it is associated Video frame figure.

Figure 31 A to 31M is to illustrate that the various aspects of technology described in displaying according to the present invention execute the synthesis of sound field Or the figure of curve graph 400A to the 400M of the various analog results of record sort.In the example of Figure 31 A to 31M, curve graph Each of 400A to 400M includes that threshold value 402 represented by dotted lines and the respective audio object 404A being represented by the dotted line are arrived 404M (is referred to as " audio object 404 ").

When determining audio object 404 lower than threshold value 402 by the analysis described above in relation to content analysis unit 26 When, content analysis unit 26 determines that the correspondence one in audio object 404 indicates the audio object recorded.Such as Figure 31 B, 31D Shown in example to 31H and 31J to 31L, content analysis unit 26 determines that audio object 404B, 404D to 404H, 404J are arrived 404L is lower than threshold value 402 (at least+90% time, and usually 100% time), and therefore indicates recorded audio pair As.As shown in the example of Figure 31 A, 31C and 31I, content analysis unit 26 determines that audio object 404A, 404C and 404I exceed Threshold value 402 and therefore expression Composite tone object.

In the example of Figure 31 M, audio object 404M indicates mixed synthesis/record audio object, with some conjunctions At part (for example, being higher than threshold value 402) and some composite parts (for example, being lower than threshold value 402).In the case, content analysis Unit 26 identifies synthesis and the record part of audio object 404M, as a result, audio coding apparatus 20 generates bit stream 21 to include base In both the coded audio data of directionality and the coded audio data based on vector.

Figure 32 is the s-matrix for illustrating the technology described according to the present invention and decomposing from high-order ambiophony coefficient The figure of the curve graph 406 of singular value.As shown in Figure 32, have the non-zero singular value being worth greatly less.The Analysis of The Acoustic Fields unit of Fig. 4 44 can analyze these singular values before the nFG of US [k] the vector 33' of the rearranged sequence of determination and V [k] the vector 35' of rearranged sequence Scape (or in other words advantage) component (usually being indicated by vector).

Figure 33 A and 33B are to illustrate to show the technology described according to the present invention to the prospect component of description sound field The figure of response curve Figure 41 0A and 410B of potential impact possessed by reordering when vector is encoded.Curve graph 410A is shown To at least some of unordered (or in other words original) US [k] vector 33 encoded as a result, and curve graph 410B show The result that corresponding person in orderly US [k] vector 33' is encoded.Top in each of curve graph 410A and 410B Mistake in plot exhibits coding, wherein may only there are apparent errors at frame boundaries in curve graph 410B.Therefore, this hair The technology of reordering described in bright can promote or facilitate in other ways using old edition tone decoder to monophonic audio pair The decoding of elephant.

Figure 34 and 35 is to illustrate the identification according to the present invention that be based solely on energy to different audio object and be based on side The concept map of difference between the identification of tropism.In the example of Figure 34, the vector for showing large energy is identified as different audio Object, but regardless of directionality how.As shown in Figure 34, according to the audio object of higher-energy value positioning (marking and drawing on the y axis) Be determined as " in the foreground ", but regardless of directionality (for example, directionality quotient representation by being plotted on the x-axis) how.

Figure 35 illustrates for example according to the technology implemented by the Analysis of The Acoustic Fields unit 44 of Fig. 4 based on both directionality and energy The identification of different audio object.As shown in Figure 35, the left side that larger directionality quotient is directed towards x-axis is marked and drawed, and larger energy level is It is marked and drawed towards the top of y-axis.In this example, Analysis of The Acoustic Fields unit 44 can determine different audio object (for example, " in the foreground " Audio object) it is associated with the upper left vector data of relative orientation curve graph is plotted as.As an example, sound field point Analysis unit 44 can determine that plotting vector those of in the left upper quadrant of curve graph is associated with different audio object.

Figure 36 A to 36F be illustrate according to the present invention described in technology various aspects spherical harmonic coefficient through decomposing At least part of version projects to execute the figure of interpolation into spatial domain.Figure 36 A be illustrate in V [k] vector 35 one or The figure of projection of more persons on spherical surface 412.In the example of Figure 36 A, the different balls that each number identification projects on spherical surface are humorous Coefficient (may be associated with a row of V matrix 19' and/or column).Different colours imply the direction of different audio component, wherein The color of thin (and gradually deepening) indicates the Main way of distinct components.Audio coding apparatus 20 shown in the example of Fig. 4 The example that can execute space-time interpolation between each of red dot to generate Figure 36 A of space-time interpolation unit 50 Shown in spherical surface.

Figure 36 B is the figure for illustrating one or more of V [k] vector 35 to the projection on cross bar.Space-time interpolation unit The example of one row and/or column of 50 projectable V [k] vectors 35 or multiple rows of V [k] vector 35 and/or column to generate Figure 36 B Shown in cross bar 414.

Figure 36 C is to illustrate one or more vectors of one or more of V [k] vector 35 to spherical surface (such as the example of Figure 36 Shown in spherical surface 412) on projection cross section figure.

Showing in Figure 36 D to 36G can illustrate in different sound sources (people in honeybee, helicopter, electronic music and stadium) The example of the snapshot of time (in about 20 milliseconds of 1 frame) when in three-dimensional space.

Technology described in the present invention is allowed to be identified using single US [k] vector sum single V [k] vector and indicates these not With the expression of sound source.The time change of sound source is to be indicated with US [k] vector, and the spatial distribution of each sound source is by list A V [k] vector indicates.One V [k] vector can indicate width, position and the size of sound source.In addition, single V [k] vector can It is expressed as the linear combination of the humorous basis function of ball.In the drawing of Figure 36 D to 36G, the expression of sound source be based on by single V to Quantitative change is changed in space coordinates.The similar approach of operation instruction sound source in Figure 36 to 36C.

Figure 37 illustrates the expression of the technology for obtaining space-time interpolation as described herein.Institute in the example of Fig. 4 The executable space-time interpolation more fully hereinafter described of the space-time interpolation unit 50 of the audio coding apparatus 20 shown. Space-time interpolation may include the acquisition higher resolution spatial component in room and time dimension the two.Spatial component can base Just in the multidimensional signal being made of higher-order ambiophony (HOA) coefficient (or HOA coefficient is referred to as " spherical harmonic coefficient ") It hands over and decomposes.

In illustrated curve graph, vector V1And V2Indicate the correspondence vector of two different spaces components of multidimensional signal. Spatial component can be decomposed by the block-by-block of multidimensional signal and be obtained.In some instances, spatial component is by relative to higher (wherein this ambiophony audio data includes the multi-pass of block, sample or any other form to rank ambiophony (HOA) audio data Audio data) each piece (it can be referred to frame) execute block-by-block form SVD and obtain.Variable M can be used to indicate audio frame Length (with number of samples).

Therefore, V1And V2It can indicate prospect V [k] vector 51 of the sequence block of HOA coefficient 11kWith prospect V [k-1] vector 51k-1Correspondence vector.V1It can for example indicate prospect V [k-1] vector 51 for first frame (k-1)k-1Primary vector, and V2 It can indicate prospect V [k] vector 51 for second and subsequent frame (k)kPrimary vector.V1And V2It can indicate that being included in multidimensional believes The spatial component of single audio object in number.

Interpolated vector V for every xxIt is by according to multidimensional signal (interpolated vector VxIt can be applied to the multidimensional Signal is to smooth time (and therefore, in some cases, space) component) time component time slice or " time sample Originally number x " is to V1And V2It is weighted and obtains.As described above, it is formed using SVD, it can be by each time samples Vector (for example, sample of HOA coefficient 11) and corresponding interpolated VxVector division is carried out to obtain the smooth of nFG signal 49.That is, US [n]=HOA [n] * Vx[n]-1, wherein this expression row vector is multiplied by column vector, thus generates the scaling element of US.Vx[n]-1It can As VxThe pseudoinverse of [n] and obtain.

Relative to V1And V2Weighting, due in time in V1The V occurred later2, V1Weight along time dimension It is lower in proportion.That is, although prospect V [k-1] vector 51k-1For the spatial component of decomposition, but prospect V [k] continuous in time to Amount 51kThe different value of representation space component over time.Therefore, V1Weight reduce, and V2Weight increase along t with x And increase.Herein, d1And d2Indicate weight.

Figure 38 is the artificial US matrix (US for illustrating the sequentially SVD block for multidimensional signal according to the techniques described herein1 And US2) block diagram.Interpolated V vector can be applied to the row vector of artificial US matrix to restore original multi-dimensional signal.It is more specific next Say, space-time interpolation unit 50 can by the pseudoinverse of interpolated prospect V [k] vector 53 multiplied by nFG signal 49 and prospect V [k] to Amount 51kThe multiplied result of (it is represented by prospect HOA coefficient) can replace the K/ of nFG signal to obtain the interpolated sample of K/2 2 samples are used as the first K/2 sample, such as U2Shown in the example of Figure 38 of matrix.

Figure 39 be illustrate according to the present invention described in technology use the smooth of singular value decomposition and spatial temporal components Come decompose higher-order ambiophony (HOA) signal subsequent frame block diagram.(it can also be expressed as frame n and frame n by frame n-1 and frame n + 1) frame continuous in time is indicated, wherein each frame includes 1024 time slices and has HOA order 4, to obtain (4+ 1)2=25 coefficients.It is artificial smooth the warp at for frame n-1 and frame n can be obtained by the interpolated V vector of application as described The US matrix of U matrix.Each grey row or column vector indicates an audio object.

The HOA of the signal based on vector is indicated in calculating effect

By take each of signal based on vector being shown in XVECk and by its it is corresponding (dequantized) Space vector VVECk is multiplied and generates instantaneous CVECk.Every VVECk is shown in MVECk.Thus, for L rank HOA signal and The M signals based on vector, will be present the M signals based on vector, and each of these person will have the dimension provided by frame length P Degree.Therefore these signals can indicate are as follows: XVECkmn, n=0 ..P-1;M=0 ..M-1.Accordingly, will be present M space to Amount, dimension (L+1)2VVECk.These are represented by MVECkml, l=0 .., (L+1)2-l;M=0 .., M-1.It is each to be based on The HOA of the signal of vector indicates that CVECkm is the matrix-vector multiplication provided as follows:

CVECkm=(XVECkm (MVECkm) T) T

It generates matrix (L+1)2Multiply P.It is provided completely and the contribution as follows to each signal based on vector is summed HOA is indicated:

CVECk=m=0M-1CVECk [m]

The space-time interpolation of V vector

However, only carrying out above calculate for the part P-B of frame length to maintain smooth space-time continuity. It is changed to by using interpolated derived from current MVECkm and preceding value MVECk-1m set MVECkml (m=0 .., M-1;l =0 .., (L+1)2) carry out HOA matrix preceding B sample.This leads to higher Time Density space vector, because we are as follows One vector is exported for each time samples p:

MVECkmp=pB-1MVECkm+B-1-pB-1MVECk-1m, p=0 .., B-1.

For each time samples p, have (L+1)2The new HOA vector of a dimension calculates are as follows:

CVECkp=(XVECkmp) MVECkmp, p=0 .., B-1

Enhance these preceding B samples by the P-B sample of previous section to cause m based on the complete of the signal of vector HOA indicates CVECkm.

At decoder (for example, audio decoding apparatus 24 shown in the example of Fig. 5), for certain different, prospects or Linear (or non-linear) interpolation can be used to carry out V vector of the interpolation from former frame and from current for advantage sound based on vector The V vector of frame is to generate the interpolated V vector of high-resolution (in time) in specific time segment.Space time interpolation list This interpolation can be performed in member 76, and wherein space-time interpolation unit 76 can be then by the US vector and high-resolution in present frame Interpolated V multiplication of vectors is to generate the HOA matrix in the specific time segment.

Alternatively, space-time interpolation unit 76 can be by the V multiplication of vectors of US vector and present frame to generate the first HOA square Battle array.In addition, decoder can be by US vector and the V multiplication of vectors from former frame to generate the 2nd HOA matrix.In space-time Insert the first HOA matrix and the 2nd HOA that linear (or non-linear) interpolation can then be applied in specific time segment by unit 76 Matrix.It is assumed that common input matrix/vector, the output of this interpolation can match the output of the multiplication of US vector and interpolated V vector.

In in this respect, the technology aloows audio coding apparatus 20 and/or audio decoding apparatus 24 to be configured To be operated according to following clause.

Clause 135054-1C.A kind of device, such as audio coding apparatus 20 or audio decoding apparatus 24 comprising: one or Multiple processors are configured in space and time the two and obtain multiple higher resolution spatial components, wherein the sky Between component be the Orthogonal Decomposition based on the multidimensional signal being made of spherical harmonic coefficient.

Clause 135054-1D.A kind of device, such as audio coding apparatus 20 or audio decoding apparatus 24 comprising: one or Multiple processors are configured to spatial component and the time point of a spherical harmonic coefficient more than smooth first and more than second a spherical harmonic coefficients At least one of amount.

Clause 135054-1E.A kind of device, such as audio coding apparatus 20 or audio decoding apparatus 24 comprising: one or Multiple processors are configured in space and time the two and obtain multiple higher resolution spatial components, wherein the sky Between component be the Orthogonal Decomposition based on the multidimensional signal being made of spherical harmonic coefficient.

Clause 135054-1G.A kind of device, such as audio coding apparatus 20 or audio decoding apparatus 24 comprising: one or Multiple processors are configured to decompose and more than second at least partially through first relative to more than first a spherical harmonic coefficients The second of spherical harmonic coefficient decomposes the spherical harmonic coefficient for increasing resolution ratio through decomposing for increasing resolution ratio and obtaining for a time slice.

Clause 135054-2G.The device of clause 135054-1G, wherein described first decomposes including indicating described more than first First V matrix of the right singular vector of a spherical harmonic coefficient.

Clause 135054-3G.The device of clause 135054-1G, wherein described second decomposes including indicating described more than second 2nd V matrix of the right singular vector of a spherical harmonic coefficient.

Clause 135054-4G.The device of clause 135054-1G, wherein described first decomposes including indicating described more than first First V matrix of the right singular vector of a spherical harmonic coefficient, and wherein described second decompose including indicating that more than second a ball is humorous 2nd V matrix of the right singular vector of coefficient.

Clause 135054-5G.The device of clause 135054-1G, wherein the time slice includes the subframe of audio frame.

Clause 135054-6G.The device of clause 135054-1G, wherein the time slice includes the time sample of audio frame This.

Clause 135054-7G.The device of clause 135054-1G, wherein one or more described processors are configured to obtain First of spherical harmonic coefficient more than described first in a spherical harmonic coefficient is decomposed and the interpolated decomposition of the second decomposition.

Clause 135054-8G.The device of clause 135054-1G, wherein one or more described processors are configured to obtain It decomposes and to contain the first of the first part of a spherical harmonic coefficient more than described first in the first frame to contain second The interpolated decomposition that the second of the second part of more than described second a spherical harmonic coefficients in frame is decomposed, wherein described at one or more Reason device is further configured to for interpolated decompose being applied to comprising a spherical harmonic coefficient more than described first in the first frame First part first time component to generate first manual time's component of a spherical harmonic coefficient more than described first, and will be corresponding It is interpolated to decompose the second time component for being applied to the second part comprising a spherical harmonic coefficient more than described second in the second frame To generate second manual time's component of more than the described second a spherical harmonic coefficients for being included.

Clause 135054-9G.The device of clause 135054-8G, wherein the first time component is by relative to institute A spherical harmonic coefficient more than first is stated to execute the synthesis based on vector and generate.

Clause 135054-10G.The device of clause 135054-8G, wherein the second time component is by relative to institute A spherical harmonic coefficient more than second is stated to execute the synthesis based on vector and generate.

Clause 135054-11G.The device of clause 135054-8G, wherein one or more described processors are further configured With: receive the first manual time component and the second manual time component;It calculates and is used for more than described first a humorous systems of ball First decomposition of several first parts and second of the second part for more than described second a spherical harmonic coefficients are decomposed interpolated It decomposes;And by the interpolated decomposition it is inverse be applied to the first manual time component with restore the first time component and Applied to the second manual time component to restore the second time component.

Clause 135054-12G.The device of clause 135054-1G, wherein one or more described processors are configured to interpolation The second space component of first spatial component of a spherical harmonic coefficient more than described first and more than second a spherical harmonic coefficient.

Clause 135054-13G.The device of clause 135054-12G, wherein first spatial component includes described in expression First U matrix of the left singular vector of a spherical harmonic coefficient more than first.

Clause 135054-14G.The device of clause 135054-12G, wherein the second space component includes described in expression 2nd U matrix of the left singular vector of a spherical harmonic coefficient more than second.

Clause 135054-15G.The device of clause 135054-12G, wherein first spatial component is indicated for described M time slice of the spherical harmonic coefficient of a spherical harmonic coefficient more than first, and the second space representation in components is used for more than described second M time slice of the spherical harmonic coefficient of a spherical harmonic coefficient.

Clause 135054-16G.The device of clause 135054-12G, wherein first spatial component is indicated for described M time slice of the spherical harmonic coefficient of a spherical harmonic coefficient more than first, and the second space representation in components is used for more than described second M time slice of the spherical harmonic coefficient of a spherical harmonic coefficient, and wherein one or more described processors are configured to obtain for institute That states time slice is decomposed last N number of element that interpolated spherical harmonic coefficient includes the first spatial component described in interpolation and described the The top n element of two spatial components.

Clause 135054-17G.The device of clause 135054-1G, wherein a spherical harmonic coefficient exists in the time domain more than described second More than described first after a spherical harmonic coefficient.

Clause 135054-18G.The device of clause 135054-1G, wherein one or more described processors are further configured It generates described first of a spherical harmonic coefficient more than described first to decompose a spherical harmonic coefficient more than described first and decomposes.

Clause 135054-19G.The device of clause 135054-1G, wherein one or more described processors are further configured It generates described second of a spherical harmonic coefficient more than described second to decompose a spherical harmonic coefficient more than described second and decomposes.

Clause 135054-20G.The device of clause 135054-1G, wherein one or more described processors are further configured To execute singular value decomposition relative to more than described first a spherical harmonic coefficients to generate the left side for indicating a spherical harmonic coefficient more than described first The U matrix of singular vector, the s-matrix for indicating more than first a spherical harmonic coefficient singular value and expression more than first a humorous system of ball The V matrix of several right singular vectors.

Clause 135054-21G.The device of clause 135054-1G, wherein one or more described processors are further configured To execute singular value decomposition relative to more than described second a spherical harmonic coefficients to generate the left side for indicating a spherical harmonic coefficient more than described second The U matrix of singular vector, the s-matrix of the singular value of expression more than second a spherical harmonic coefficient and expression more than second a ball are humorous The V matrix of the right singular vector of coefficient.

Clause 135054-22G.The device of clause 135054-1G, wherein first and second described multiple spherical harmonic coefficient are respectively Indicate that the plane wave of sound field indicates.

Clause 135054-23G.The device of clause 135054-1G, wherein first and second described multiple spherical harmonic coefficient are respectively Indicate one or more the monophonic audio objects mixed.

Clause 135054-24G.The device of clause 135054-1G, wherein first and second described multiple spherical harmonic coefficient are respectively First and second corresponding spherical harmonic coefficient comprising indicating three-dimensional sound field.

Clause 135054-25G.The device of clause 135054-1G, wherein first and second described multiple spherical harmonic coefficient are respectively Join at least one spherical substrate functional dependence with the order greater than one.

Clause 135054-26G.The device of clause 135054-1G, wherein first and second described multiple spherical harmonic coefficient are respectively Join at least one spherical substrate functional dependence with the order equal to four.

Clause 135054-27G.The device of clause 135054-1G, wherein the interpolation is first decomposition and second point The weighted interpolation of solution, wherein the weight for being applied to the weighted interpolation that described first decomposes is decomposed with by described first and second Vector indicate time be inversely proportional, and wherein be applied to it is described second decompose weighted interpolation weight with by described first And second decompose vector indicate time it is proportional.

Clause 135054-28G.The device of clause 135054-1G, wherein described through decomposing the interpolated smooth institute of spherical harmonic coefficient State at least one of spatial component and the time component of a spherical harmonic coefficient more than first and more than second a spherical harmonic coefficient.

Figure 40 A to 40J, which is respectively, illustrates the various aspects that technology described in the present invention can be performed with packed description two The block diagram of example audio code device 510A to the 510J of the spherical harmonic coefficient of dimension or three-dimensional sound field.In the example of Figure 40 A to 40J Each in, in some instances, audio coding apparatus 510A and 510B respectively indicate that audio data can be encoded Any device, such as it is desktop PC, laptop computer, work station, plate (tablet, slate) computer, dedicated Audio recording device, cellular phone (including so-called " smart phone "), personal media player apparatus, individual tour theatrical costume It sets, or the device for any other type that audio data can be encoded.

Although being shown as single device, that is, device 510A to the 510J in the example of Figure 40 A to 40J is hereinafter referred to as wrapped Contained in device 510A to 510J various assemblies or unit can actually form the independent dress outside device 510A to 510J It sets.In other words, although be described as in the present invention by single device (that is, device 510A in the example of Figure 40 A to 40J to It 510J) executes, but the technology can be implemented by the system including multiple devices or in addition be executed, it is wherein each in these devices Person can respectively contain one or more of the various assemblies being described more particularly below or unit.Therefore, the technology should not limit In the example of Figure 40 A to 40J.

In some instances, audio coding apparatus 510A to 510J is indicated described in the example above in relation to Fig. 3 and 4 The substitution of audio coding apparatus.Through the following discussion to audio coding apparatus 510A to 510J, for above in relation to Fig. 4 institute The various units 30 to 52 of the audio coding apparatus 20 of description point out various similitudes operationally.In many aspects, such as Lower described, the mode that audio coding apparatus 510A to 510J can be substantially similar to audio coding apparatus 20 operates, but has Slight deviations or modification.

As shown in the example of Figure 40 A, audio coding apparatus 510A includes audio compressing unit 512, audio coding unit 514 and bitstream producing unit 516.Audio compressing unit 512 can indicate the list of compression spherical harmonic coefficient (SHC) 511 (" SHC 511 ") Member, SHC 511 can also be expressed as higher-order ambiophony (HOA) coefficient 511.Audio compressing unit 512 can be in some cases Under, audio compressing unit 512 indicates the unit that can nondestructively compress SHC 511 or execute lossy compression relative to SHC 511. SHC 511 can indicate multiple SHC, wherein at least one of the multiple SHC corresponds to the spherical substrate letter that order is greater than one Number (wherein have this multifarious SHC referred to as higher-order ambiophony (HOA) mutually to distinguish with lower-order ambiophony, One example of lower-order ambiophony is so-called " B format "), such as it is described in more detail above.Although audio compressing unit 512 can nondestructively compress SHC 511, but in some instances, and the removal of audio compressing unit 512 SHC's 511 is describing sound field Those not prominent or incoherent SHC (because some SHC may not be heard by people's auditory system) when (in regeneration).? In this meaning, the property that damages of this compression may not be in excessive influence sound field when regenerating through compressed version from SHC 511 Perceived quality.

In the example of Figure 40 A, audio compressing unit includes 518 harmony field component extraction unit 520 of decomposition unit.It decomposes Unit 518 can be similar to the Linear Invertible Transforms unit 30 of audio coding apparatus 20.That is, decomposition unit 518 can indicate to be configured The unit of analysis in the form of executing and be known as singular value decomposition.It, can be linear not relative to offer although being described relative to SVD The execution technology is decomposed in any similar transformation of the set of related data.Also, in the present invention to the reference of " set " Set reference " non-zero " gathers (unless specifically state otherwise), and not set reference includes the set of so-called " null set " Classical mathematics definition.

Under any circumstance, (it again can be by its acronym " SVD " for the execution of decomposition unit 518 singular value decomposition Indicate) with by spherical harmonic coefficient 511 be transformed into transformed spherical harmonic coefficient two or more set.In the example of Figure 40, Decomposition unit 518 can execute SVD relative to SHC 511 to generate so-called V matrix 519, s-matrix 519B and U matrix 519C.? In the example of Figure 40, decomposition unit 518 individually each of output matrix rather than such as above in relation to Linear Invertible Transforms Unit 30 discusses output US [k] vector in combination.

As noted above, the V* matrix in above-mentioned SVD mathematic(al) representation be expressed as the conjugate transposition of V matrix with Reflection SVD can be applied to include plural matrix.When being applied to only includes the matrix of real number, the complex conjugate of V matrix (or is changed Sentence is talked about, V* matrix) it is believed that being equal to V matrix.Hereinafter it is easy the purpose of explanation, it is assumed that SHC 511 includes real number, as a result Via SVD rather than V* Output matrix V matrix.Although it is assumed that V matrix, the technology can be applied to have multiple in a similar manner The SHC 511 of number system number, wherein the output of SVD is V* matrix.Therefore, in this regard, the technology should not necessarily be limited by only to provide and answer It with SVD to generate V matrix, and may include that SVD is applied to the SHC 511 with complex number components to generate V* matrix.

Under any circumstance, decomposition unit 518 can (wherein this be vertical relative to higher-order ambiophony (HOA) audio data Volume reverberation audio data includes the block or sample of the multichannel audio data of SHC 511 or any other form) each piece of (its Can be referred to as frame) execute block-by-block form SVD.Variable M can be used to indicate the length of audio frame (with number of samples).For example, When audio frame includes 1024 audio samples, M is equal to 1024.Decomposition unit 518 can be therefore relative to multiplying (N+1) with M2It is a The block of the SHC 511 of SHC executes block-by-block SVD, and wherein N indicates the order of HOA audio data again.Decomposition unit 518 can pass through It executes this SVD and generates V matrix 519, s-matrix 519B and U matrix 519C, wherein in matrix 519 to 519C (" matrix 519 ") Each can indicate V, S and U matrix corresponding in greater detail above.Decomposition unit 518 can by these matrixes 519A transmit or It is output to sound field component extraction unit 520.V matrix 519A can have size (N+1)2Multiply (N+1)2, s-matrix 519B can have greatly Small (N+1)2Multiply (N+1)2, and U matrix can have size M to multiply (N+1)2, wherein M refers to the number of the sample in audio frame.The allusion quotation of M Offset is 1024, but technology of the invention should not necessarily be limited by this representative value of M.

Sound field component extraction unit 520 can indicate the distinct components for being configured to determine and then extracting sound field and sound field Unit of the background component to effectively separate the distinct components of sound field with the background component of sound field.In in this respect, sound field The Analysis of The Acoustic Fields unit of audio coding apparatus 20 shown in the executable example above in relation to Fig. 4 of component extraction unit 520 44, many persons in the described operation of Foreground selection unit 48 and foreground selection unit 36.It is assumed that in some instances, sound field Distinct components need higher-order (background component relative to sound field) basis function (and therefore needing more SHC) to come accurately Distinct components, are separated with background component that may make can be by more bit allocation to different point by the different property for indicating these components It measures and less bits (in contrast) is distributed into background component.Therefore, application this transformation (in the form of SVD or it is any its Its variation includes PCA), technology described in the present invention can promote bit allocation to various SHC, and compress SHC in turn 511。

In addition, as follows be more fully described relative to Figure 40 B, the technology can also be achieved the background component of sound field Degree reduction, it is assumed that do not need higher-order basis function in some instances to indicate that these background parts of sound field (give this The diffusion or background property of a little components).Therefore, the diffusion or the compression in terms of background that the technology can realize sound field are led to simultaneously Cross the distinct components outstanding or aspect that SVD is applied to SHC 511 and retains sound field.

As further shown in the example of Figure 40, sound field component extraction unit 520 includes transposition unit 522, prominent component Analytical unit 524 and mathematical units 526.The expression of transposition unit 522 is configured to transposition V matrix 519A to generate V matrix 519 (it is expressed as " V to transpositionTMatrix 523 ") unit.Transposition unit 522 can be by this VTMatrix 523 is output to mathematical units 526.VT Matrix 523 can have size (N+1)2Multiply (N+1)2

Prominent Component Analysis unit 524 indicates to be configured to the unit that high-lighting analysis is executed relative to s-matrix 519B.? In this respect, audio coding dress shown in the prominent executable example being similar to above in relation to Fig. 4 of Component Analysis unit 524 Set the 20 described operation operated of Analysis of The Acoustic Fields unit 44.Prominent Component Analysis unit 524 can analyze pair of s-matrix 519B Linea angulata value, to select variable these components of D number with maximum value.In other words, prominent Component Analysis unit 524 can Determine two sub-spaces of separation (for example, prospect or advantage by the slope of a curve of the diagonal line value generation of successively decreasing of S by analyzing Subspace and background or environment subspace) value D, wherein big singular value indicates prospect or different sound, and low singular value indicates The background component of sound field.In some instances, the single order of singular value curve and second order can be used to lead for prominent Component Analysis unit 524 Number.Number D can be also limited between one and five by prominent Component Analysis unit 524.As another example, prominent Component Analysis list Number D can be limited in one and (N+1) by member 5242Between.Alternatively, prominent Component Analysis unit 524 can predefined number D, so as to Reach value four.Under any circumstance, once estimating number D, prominent Component Analysis unit 24 is just before matrix U, V and S extraction Scape and background subspace.

In some instances, prominent Component Analysis unit 524 can every M sample (it can be restated as frame by frame) execution This analysis.In in this respect, D can be in interframe difference.In other examples, prominent Component Analysis unit 24 can every frame execute this point It analyses more than once, thus two or more parts of analysis frame.Therefore, the technology should not necessarily be limited by the present invention in this respect Described in example.

In fact, prominent Component Analysis unit 524 can analyze diagonal matrix, (it is expressed as S square in the example of Figure 40 Battle array 519B) singular value, other values those of value of the identification relative value greater than diagonal line s-matrix 519B.Prominent Component Analysis list First 524 recognizable D values, extract these values to generate SDISTMatrix 525A and SBGMatrix 525B.SDISTMatrix 525A can indicate to wrap Include (N+1) with original s-matrix 519B2D column diagonal matrix.In some cases, SBGMatrix 525B can indicate have There is (N+1)2The matrix of-D column, each of these person include (N+1) of original s-matrix 519B2A transformed spherical harmonic coefficient.Although It is described as indicating that the D column including original s-matrix 519B (have (N+1)2A value) matrix, but prominent Component Analysis unit 524 This matrix can be truncated to generate the S of the D column (with D value) with original s-matrix 519BDISTMatrix, it is assumed that s-matrix 519B is (N+1) of D column after diagonal matrix and D value in each column2Value is usually from zero.Although relative to complete SDIST Matrix 525A and complete SBGMatrix 525B description, but can be relative to these SDISTThe truncated version of matrix 525A and this SBGMatrix The truncated version of 525B implements the technology.Therefore, in this regard, technology of the invention should not be restricted.

In other words, SDISTMatrix 525A can have size D to multiply (N+1)2, and SBGMatrix 525B can have size (N+1)2- D multiplies (N+1)2。SDISTMatrix 525A may include being confirmed as those of protrusion in terms of different (DIST) audio component for sound field Principal component or (in other words) singular value, and SBGMatrix 525B may include background (BG) or (in other words) for being confirmed as sound field Those of environment or non-different audio component singular value.Although being shown as independent matrix 525A and 525B in the example of Figure 40, But matrix 525A and 525B can be designed to single matrix, indicate that this indicates S using variables DDISTThe single matrix of matrix 525 Arrange the number of (from left to right).In some instances, variables D may be set to four.

Prominent Component Analysis unit 524 can also analyze U matrix 519C to generate UDISTMatrix 525C and UBGMatrix 525D.Often Often, prominent Component Analysis unit 524 can analyze s-matrix 519B to identify variables D, to generate U based on variables DDISTMatrix 525C and UBGMatrix 525B.That is, prominent Component Analysis unit 524 can base after the D outstanding column for identifying s-matrix 519B U matrix 519C is divided in the variables D that this is determined.In the case, prominent Component Analysis unit 524 can produce UDISTMatrix 525C With (N+1) comprising original U matrix 519C2The D of a transformed spherical harmonic coefficient arranges (from left to right), and generates UBGMatrix 525D with (N+1) comprising original U matrix 519C2The residue (N+1) of a transformed spherical harmonic coefficient2- D column.UDISTMatrix 525C can have greatly Small M multiplies D, and UBGMatrix 525D can have size M to multiply (N+1)2-D.Although being shown as independent matrix 525C in the example of Figure 40 And 525D, but matrix 525C and 525D can be designed to single matrix, indicate that this indicates U using variables DDISTThe list of matrix 525B The number of a matrix column (from left to right).

Prominent Component Analysis unit 524 can also analyze VTMatrix 523 is to generate VT DISTMatrix 525E and VT BGMatrix 525F. Usually, prominent Component Analysis unit 524 can analyze s-matrix 519B to identify variables D, to generate V based on variables DT DISTMatrix 525E and VBGMatrix 525F.That is, prominent Component Analysis unit 524 can base after the D outstanding column for identifying s-matrix 519B V matrix 519A is divided in the variables D that this is determined.In the case, prominent Component Analysis unit 524 can produce VT DISTMatrix 525E To include original VTMatrix 523 is (N+1) of D value2Row (from top to bottom), and generate VT BGMatrix 525F is to include original VTSquare Battle array 523 is (N+1)2The residue (N+1) of-D value2Row.VT DISTMatrix 525E can have size (N+1)2Multiply D, and VT BGMatrix 525D can have size (N+1)2Multiply (N+1)2-D.Although being shown as independent matrix 525E and 525F, square in the example of Figure 40 Battle array 525E and 525F can be designed to single matrix, indicate that this indicates V using variables DDISTThe single matrix column of matrix 525E (from left to right) number.Prominent Component Analysis unit 524 can be by SDISTMatrix 525, SBGMatrix 525B, UDISTMatrix 525C, UBGMatrix 525D and VT BGMatrix 525F is output to mathematical units 526, while also by VT DISTMatrix 525E is output to bit stream generation Unit 516.

Mathematical units 526 can indicate to be configured to execute the matrix that can be executed relative to one or more matrixes (or vector) Multiplication or the unit of any other mathematical operation.More specifically, as shown in the example of Figure 40, mathematical units 526 can be indicated It is configured to execute matrix multiplication with by UDISTMatrix 525C is multiplied by SDISTMatrix 525A multiplies the U of D to generate with size MDIST* SDISTThe unit of vector 527.Matrix mathematical units 526 can also indicate to be configured to execute matrix multiplication with by UBGMatrix 525D multiplies With SBGMatrix 525B and then multiplied by VT BGMatrix 525F is to generate UBG*SBG*VT BGMatrix 525F is multiplied using generating size as size M (N+1)2Background spherical harmonic coefficient 531 (it can indicate the humorous system of those of the background component of expression sound field of spherical harmonic coefficient 511 ball Number) unit.Mathematical units 526 can be by UDIST*SDISTVector 527 and background spherical harmonic coefficient 531 are output to audio coding unit 514。

Therefore, audio coding apparatus 510 is different from audio coding apparatus 20, the difference is that audio coding apparatus 510 Matrix multiplication at the end of comprising being configured to pass cataloged procedure and generate UDIST*SDISTVector 527 and background spherical harmonic coefficient 531 this mathematical units 526.The Linear Invertible Transforms unit 30 of audio coding apparatus 20 executes the multiplication of U and s-matrix to compile Opposite beginning output US [k] vector 33 of code process, can promote and do not operated shown in the example of Figure 40 later, such as It reorders.In addition, audio coding apparatus 20 restores background SHC 531 not at the end of cataloged procedure, but directly from HOA system Number 11 selects background HOA coefficient 47, and then avoids matrix multiplication potentially to restore background SHC 531.

Audio coding unit 514 can indicate to execute some form of coding further to compress UDIST*SDISTVector 527 and back The unit of scape spherical harmonic coefficient 531.Audio coding unit 514 can be substantially similar to audio coding shown in the example of Fig. 4 The mode of the psychologic acoustics tone decoder unit 40 of device 20 operates.In some cases, this audio coding unit 514 can table Show one or more examples of Advanced Audio Coding (AAC) coding unit.Audio coding unit 514 can be to UDIST*SDISTVector 527 Each column or row encoded.Usually, audio coding unit 514 can be for order/son remaining in background spherical harmonic coefficient 531 The example that AAC coding unit is called in each of rank combination.About can how using AAC coding unit to background spherical harmonic coefficient 531 more information encoded are found in the entitled of Eric He Lude (Eric Hellerud) et al. and " are compiled with AAC In the conference paper of code higher-order ambiophony (Encoding Higher Order Ambisonics with AAC) ", It submits and can be obtained at lower place in 124th conference (17 to 20 May in 2008):http://ro.uow.edu.au/ Cgi/viewcontent.cgi? article=:8025&context=engpapers.Audio coding unit 14 can be by UDIST* SDISTThe encoded version of vector 527 (is expressed as " encoded UDIST*SDISTVector 515 ") and background spherical harmonic coefficient 531 warp knit Code version (being expressed as " encoded background spherical harmonic coefficient 515B ") is output to bitstream producing unit 516.In some cases, audio Coding unit 514 can be used ratio to UDIST*SDISTThe low bit rate of the bit rate that vector 527 is encoded is humorous to background ball Coefficient 531 carry out audio coding, and then potentially with UDIST*SDISTVector 527 is compared to more compressed background spherical harmonic coefficient 531.

516 presentation format data of bitstream producing unit are to meet known format (it can refer to format known to decoding apparatus) And then generate the unit of bit stream 517.Bitstream producing unit 42 can be substantially similar to shown in the example above in relation to Fig. 4 Audio coding device 24 bitstream producing unit 42 described in mode operate.Bitstream producing unit 516 may include to warp knit Code UDIST*SDISTVector 515, encoded background spherical harmonic coefficient 515B and VT DISTThe multiplexing that matrix 525E is multiplexed Device.

Figure 40 B is that the various aspects of technology described in the present invention can be performed with packed description two dimension or three dimensional sound in explanation The block diagram of the example audio code device 510B of the spherical harmonic coefficient of field.Audio coding apparatus 510B can be similar to audio coding apparatus 510, similar place is that audio coding apparatus 510B includes that audio compressing unit 512, audio coding unit 514 and bit stream generate Unit 516.In addition, the audio compressing unit 512 of audio coding apparatus 510B can be similar to the audio pressure of audio coding apparatus 510 Contracting unit, similar place are that audio compressing unit 512 includes decomposition unit 518.The audio compression of audio coding apparatus 510B Unit 512 may differ from the audio compressing unit 512 of audio coding apparatus 510, the difference is that sound field component extraction list Additional unit of the member 520 comprising being expressed as Degree reduction unit 528A (" Degree reduction unit 528 ").For this reason, audio The sound field component extraction unit 520 of code device 510B is expressed as " sound field component extraction unit 520B ".

Degree reduction unit 528A indicates the unit for being configured to execute the additional Degree reduction of background spherical harmonic coefficient 531. In some cases, sound field represented by the rotatable background spherical harmonic coefficient 531 of Degree reduction unit 528A indicates sound field to reduce The number of necessary background spherical harmonic coefficient 531.In some cases, it is assumed that the background of the expression sound field of background spherical harmonic coefficient 531 Component, Degree reduction unit 528A can be removed, eliminate or delete in other ways (often through pulverised) background spherical harmonic coefficient 531 In correspond to those of higher-order spherical substrate function spherical harmonic coefficient.In in this respect, class is can be performed in Degree reduction unit 528A It is similar to the operation of the Foreground selection unit 48 of audio coding apparatus 20 shown in the example of Fig. 4.Degree reduction unit 528A can Background spherical harmonic coefficient 531 is output to audio coding list through reduction version (being expressed as " through reducing background spherical harmonic coefficient 529 ") Member 514, audio coding unit 514 can execute audio coding to through reducing background spherical harmonic coefficient in a manner of as described above 529 carry out coding and generate in turn encoded through reducing background spherical harmonic coefficient 515B.

The various aspects of technology described in the present invention can be presented in the various clauses being listed below.

Clause 132567-1.A kind of device, such as audio coding apparatus 510 or audio coding apparatus 510B comprising one Or multiple processors, one or more described processors are configured to execute following operation: humorous relative to the multiple balls for indicating sound field Coefficient executes singular value decomposition to generate the U matrix for the left singular vector for indicating the multiple spherical harmonic coefficient, indicate the multiple The V matrix of the right singular vector of the s-matrix and the multiple spherical harmonic coefficient of expression of the singular value of spherical harmonic coefficient;And it will be the multiple Spherical harmonic coefficient is expressed as at least part of function of one or more of U matrix, s-matrix and V matrix.

Clause 132567-2.The device of clause 132567-1, wherein one or more described processors are further configured to Generate bit stream with by the multiple spherical harmonic coefficient indicate be included as U matrix, s-matrix and V matrix one or more vectors, include A combination thereof or its export item.

Clause 132567-3.The device of clause 132567-1, wherein one or more described processors are further configured to One or more of the distinct components for the description sound field being contained in the U matrix are determined when indicating the multiple spherical harmonic coefficient UDISTVector.

Clause 132567-4.The device of clause 132567-1, wherein one or more described processors are further configured to One or more of the distinct components for the description sound field being contained in the U matrix are determined when indicating the multiple spherical harmonic coefficient UDISTVector determines one or more S for the distinct components for also describing sound field being contained in the s-matrixDISTVector, and by institute State one or more UDISTVector and one or more described one or more SDISTMultiplication of vectors is to generate UDIST*SDISTVector.

Clause 132567-5.The device of clause 132567-1, wherein one or more described processors are further configured to One or more of the distinct components for the description sound field being contained in the U matrix are determined when indicating the multiple spherical harmonic coefficient UDISTVector determines one or more S for the distinct components for also describing sound field being contained in the s-matrixDISTVector, and by institute State one or more UDISTVector and one or more described one or more SDISTMultiplication of vectors is to generate UDIST*SDISTVector, and wherein One or more described processors are further configured to one or more described UDIST*SDISTVector carries out audio coding to generate One or more described UDIST*SDISTThe audio coding version of vector.

Clause 132567-6.The device of clause 132567-1, wherein one or more described processors are further configured to One or more U being contained in the U matrix are determined when indicating the multiple spherical harmonic coefficientBGVector.

Clause 132567-7.The device of clause 132567-1, wherein one or more described processors are further configured to The s-matrix is analyzed when indicating the multiple spherical harmonic coefficient to identify the distinct components and background component of sound field.

Clause 132567-8.The device of clause 132567-1, wherein one or more described processors are further configured to The s-matrix is analyzed when indicating the multiple spherical harmonic coefficient to identify the distinct components and background component of sound field, and based on to S The analysis of matrix determines one or more U of the distinct components of the description sound field of the U matrixDISTVector and the U matrix are retouched State one or more U of the background component of sound fieldBGVector.

Clause 132567-9.The device of clause 132567-1, wherein one or more described processors are further configured to The s-matrix is analyzed when indicating the multiple spherical harmonic coefficient by audio frame to identify the distinct components and background point of sound field Amount, and one or more U of the distinct components based on the description sound field for analyzing determining U matrix by audio frame to the s-matrixDIST One or more U of the background component of the description sound field of vector and U matrixBGVector.

Clause 132567-10.The device of clause 132567-1, wherein one or more described processors are further configured to The s-matrix is analyzed when indicating the multiple spherical harmonic coefficient to identify the distinct components and background component of sound field, and based on to S The analysis of matrix determines one or more U of the distinct components of the description sound field of U matrixDISTThe description sound field of vector and U matrix One or more U of background componentBGVector determines one or more U corresponding to described in of s-matrix based on the analysis to s-matrixDISTTo Amount and one or more described UBGOne or more S of vectorDISTVector and one or more SBGVector, and based on the analysis to s-matrix Determine the transposition of V matrix corresponds to one or more described UDISTVector and one or more described UBGOne or more V of vectorT DIST Vector and one or more VT BGVector.

Clause 132567-11.The device of clause 132567-10, wherein one or more described processors are further configured Further to incite somebody to action one or more described U when indicating the multiple spherical harmonic coefficientBGVector is multiplied by one or more described SBGVector and Then multiplied by one or more VT BGVector is to generate one or more UBG*SBG*VT BGVector, and wherein one or more described processors It is further configured to the UBG*SBG*VT BGVector carries out audio coding to generate the UBG*SBG*VT BGThe audio of vector is compiled Code version.

Clause 132567-12.The device of clause 132567-10, wherein one or more described processors are further configured With one or more U by described in when indicating the multiple spherical harmonic coefficientBGVector is multiplied by one or more described SBGVector and then multiply With one or more VT BGVector is to generate one or more UBG*SBG*VT BGVector, and Degree reduction process is executed to eliminate described one Or multiple UBG*SBG*VT BGThose coefficients associated with one or more orders of the humorous basis function of ball in the coefficient of vector and And then generate one or more described UBG*SBG*VT BGThe order of vector is through reducing version.

Clause 132567-13.The device of clause 132567-10, wherein one or more described processors are further configured With one or more U by described in when indicating the multiple spherical harmonic coefficientBGVector is multiplied by one or more described SBGVector and then multiply With one or more VT BGVector is to generate one or more UBG*SBG*VT BGVector, and Degree reduction process is executed to eliminate described one Or multiple UBG*SBG*VT BGThose coefficients associated with one or more orders of the humorous basis function of ball in the coefficient of vector and And then generate one or more described UBG*SBG*VT BGThe order of vector through reducing version, and wherein one or more described processors into One step is configured to one or more described UBG*SBG*VT BGThe order of vector carries out audio coding through reduction version to generate rank Number is reduced one or more UBG*SBG*VT BGThe audio coding version of vector.

Clause 132567-14.The device of clause 132567-10, wherein one or more described processors are further configured With one or more U by described in when indicating the multiple spherical harmonic coefficientBGVector is multiplied by one or more described SBGVector and then multiply With one or more VT BGVector is to generate one or more UBG*SBG*VT BGVector, execute Degree reduction process with eliminate described one or Multiple UBG*SBG*VT BGIt is associated with one or more orders for being greater than one of humorous basis function of ball in the coefficient of vector One or more described U of those coefficients and in turn generationBG*SBG*VT BGThe order of vector through reduce version, and to it is described one or more UBG*SBG*VT BGThe order of vector carries out audio coding through reduction version and is reduced one or more U to generate orderBG*SBG*VT BGTo The audio coding version of amount.

Clause 132567-15.The device of clause 132567-10, wherein one or more described processors are further configured To generate bit stream comprising one or more described VT DISTVector.

Clause 132567-16.The device of clause 132567-10, wherein one or more described processors are further configured To generate bit stream comprising one or more described VT DISTVector is without to one or more described VT DISTVector carries out audio coding.

Clause 132567-1F.A kind of device, such as audio coding apparatus 510 or 510B, including one or more processors with The multichannel is indicated relative at least part of multichannel audio data execution singular value decomposition for indicating sound field to generate Described in the U matrix of the left singular vector of audio data, the s-matrix and expression of the singular value of the expression multichannel audio data The V matrix of the right singular vector of multichannel audio data, and the multichannel audio data is expressed as the U matrix, the S At least part of function of one or more of matrix and the V matrix.

Clause 132567-2F.The device of clause 132567-1F, wherein the multichannel audio data includes that multiple balls are humorous Coefficient.

Clause 132567-3F.The device of clause 132567-2F, wherein one or more described processors are further configured To be executed as described in any combination of clause 132567-2 to 132567-16.

From each of above-mentioned various clauses, it should be appreciated that any one of audio coding apparatus 510A to 510J can be held Row method, or also comprise the dress for executing each step that audio coding apparatus 510A to 510J is configured to the method executed It sets.In some cases, these devices may include one or more processors.In some cases, one or more described processors It can indicate by means of storing the application specific processor to the instruction configuration of non-transitory computer-readable storage media.In other words, The various aspects of technology in each of set of encoding example can provide the non-transitory calculating for being stored thereon with instruction Machine readable memory medium, described instruction cause one or more processors execution audio coding apparatus 510A to arrive when being executed 510J has been configured to the method executed.

For example, clause 132567-17 can be exported from previous clause 132567-1 and is a kind of method, the method packet Include: multiple spherical harmonic coefficients relative to expression sound field execute singular value decomposition to generate the left surprise for indicating the multiple spherical harmonic coefficient The U matrix of incorgruous amount, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and the right surprise for indicating the multiple spherical harmonic coefficient The V matrix of incorgruous amount;And the multiple spherical harmonic coefficient is expressed as one or more of U matrix, s-matrix and V matrix at least The function of a part.

As another example, clause 132567-18 can be exported from previous clause 132567-1 and is a kind of device, such as sound Frequency code device 510B comprising: for executing singular value decomposition relative to multiple spherical harmonic coefficients of expression sound field to generate table Show the s-matrix and table of the U matrix of the left singular vector of the multiple spherical harmonic coefficient, the singular value for indicating the multiple spherical harmonic coefficient Show the device of the V matrix of the right singular vector of the multiple spherical harmonic coefficient;And for the multiple spherical harmonic coefficient to be expressed as U square The device of at least part of function of one or more of battle array, s-matrix and V matrix.

As yet another embodiment, clause 132567-18 can be to be stored with above one kind from previous clause 132567-1 export The non-transitory computer-readable storage media of instruction, described instruction cause one or more processors relative to expression when being executed Multiple spherical harmonic coefficients of sound field execute singular value decomposition to generate the U square for the left singular vector for indicating the multiple spherical harmonic coefficient The V square of the right singular vector of battle array, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and the multiple spherical harmonic coefficient of expression Battle array, and the multiple spherical harmonic coefficient is expressed as at least part of letter of one or more of U matrix, s-matrix and V matrix Number.

For as illustrated and derived various devices, method and the computer-readable storage matchmaker of non-transitory above Body, various clauses can be exported similarly from clause 132567-2 to 132567-16.Can for through the present invention list it is various its Its clause executes same operation.

Figure 40 C is that the various aspects of technology described in the present invention can be performed with packed description two dimension or three dimensional sound in explanation The block diagram of the example audio code device 510C of the spherical harmonic coefficient of field.Audio coding apparatus 510C can be similar to audio coding apparatus 510B, similar place are that audio coding apparatus 510C includes that audio compressing unit 512, audio coding unit 514 and bit stream produce Raw unit 516.In addition, the audio compressing unit 512 of audio coding apparatus 510C can be similar to the sound of audio coding apparatus 510B Frequency compression unit, similar place are that audio compressing unit 512 includes decomposition unit 518.

However, the audio compressing unit 512 of audio coding apparatus 510C may differ from the audio of audio coding apparatus 510B Compression unit 512, the difference is that sound field component extraction unit 520 includes to be expressed as vector to reorder the volume of unit 532 Outer unit.For this reason, the sound field component extraction unit 520 of audio coding apparatus 510C is expressed as " sound field component extraction list First 520C ".

The vector unit 532 that reorders can indicate to be configured to UDIST*SDISTVector 527 reorders to generate rearranged sequence One or more UDIST*SDISTThe unit of vector 533.In in this respect, the vector unit 532 that reorders can be similar to phase above The mode described in unit 34 that reorders of audio coding apparatus 20 shown in example for Fig. 4 operates.Sound field component mentions Take unit 520C that vector can be called to reorder unit 532 to UDIST*SDISTVector 527 reorders, because of UDIST*SDISTVector 527 order (wherein UDIST*SDISTEach vector of vector 527 can indicate one or more the different monophonics being present in sound field Audio object) may each section due to being indicated above and for audio data it is different.That is, in some instances, it is false Determining these parts of audio compressing unit 512 to the audio data of commonly referred to as audio frame, (it can have spherical harmonic coefficient 511 M sample, wherein M is set as 1024) operating in some instances, the vector corresponding to these different monophonic audio objects Position (is such as shown in export UDIST*SDISTIn the U matrix 519C of vector 527) it can be different in audio interframe.

By these UDIST*SDISTVector 527 is directly delivered to audio coding unit 514 without to these UDIST*SDISTVector 527 by audio frame reorder may reduce some compression schemes (such as old edition compression scheme, monophonic audio object across More audio frame correlation is (by channel, in this example by UDIST*SDISTThe order of vector 527 relative to each other defines) when hold Achievable compression degree capablely more preferably).In addition, when not reordering, UDIST*SDISTThe coding of vector 527 may be when restoring Reduce the quality of audio data.For example, with direct frame by frame to UDIST*SDISTThe compression that vector 527 is realized when being encoded It compares, frame by frame compression can more effectively can be passed through by the AAC encoder that audio coding unit 514 indicates in the example of Figure 40 C One or more U to reorderDIST*SDISTVector 533.Although being described by above in relation to AAC encoder, can relative to across Frame more in specific order or position provides any coding more preferably compressed when specifying monophonic audio object (by channel) Device executes the technology.

As described in greater detail below, the technology aloow audio coding apparatus 510C reorder one or more to Amount is (that is, UDIST*SDISTVector 527 is to generate one or more vectors of rearranged sequence UDIST*SDISTVector 533 and in turn promotion pass through Old editions audio coders such as audio coding unit 514 to UDIST*SDISTThe compression of vector 527.Audio coding apparatus 510C can Execute technology described in the present invention further to use audio coding unit 514 to one or more U of rearranged sequenceDIST* SDISTVector 533 carries out audio coding to generate one or more U of rearranged sequenceDIST*SDISTThe encoded version of vector 533 515A。

For example, sound field component extraction unit 520C can call vector to reorder unit 532 to from time In one or more the 2nd UDIST*SDISTThe first audio frame after second frame corresponding to vector 527 one or more first UDIST*SDISTVector 527 reorders.Although being given in the context of the first audio frame after the second audio frame in time With description, but the first audio frame can be in time prior to the second audio frame.Therefore, the technology, which should not necessarily be limited by the present invention, describes Example.

Vector reorders unit 532 can be first relative to the first UDIST*SDISTVector 527 and the 2nd UDIST*SDISTVector Each of 527 execute energy spectrometer, calculate at least part (but usually from entire) and the second audio of the first audio frame The root mean square energy of a part (but usually from entire) of frame and in turn eight energy of generation (it is assumed that D is four), for the first sound First U of frequency frameDIST*SDISTEach of vector 527 generates one and is directed to the 2nd U of the second audio frameDIST*SDISTTo Each of amount 527 generates one.The vector unit 532 that reorders can then compare by wheel from the first UDIST*SDISTVector 527 each energy and the 2nd UDIST*SDISTEach of vector 527, as above in relation to described by table 1 to 4.

It in other words, can when being decomposed to HoA signal using the SVD (or correlation technique, such as KLT and PCA) based on frame It can not guarantee that the sequence of vector is consistent in interframe.For example, if decomposing (its there are two objects in basic sound field Can be referred to " desirable broken-down " when properly executing) separation that can lead to described two objects, so that a vector will indicate U An object in matrix.However, even if when decompose be represented by " desirable broken-down " when, vector (and accordingly exists in U matrix In S and V matrix) position may also replace in interframe.In addition, phase difference may be present, wherein vector reorder unit 532 can Make phasing back using phasing back (by by each element dot product negative one of inverted vector).In order to frame by frame by these Vector is fed to same " AAC/ audio coding engine ", it may be necessary to identify order (or in other words matched signal), correction phase Position and at frame boundaries apply careful interpolation.In the case where this condition, underlying audio codec can produce exceedingly odious Illusion, comprising be referred to as those of ' time hangover ' or ' pre- echo ' illusion.

The various aspects of technology described according to the present invention, audio coding apparatus 510C can make using multiple methods With at the frame boundaries of vector energy and crosscorrelation identify/matching vector.Audio coding apparatus 510C also can ensure that usually The phase change of the 180 degree occurred at frame boundaries is corrected.The vector unit 532 that reorders can apply certain shape between vector The fade in/out interpolation window of formula ensures the smooth transition between frame.

By this method, audio coding apparatus 530C can reorder to one or more vectors to generate the one or more of rearranged sequence A primary vector and the coding for promoting old edition audio coder in turn, wherein one or more described vector descriptions indicate the phase of sound field Different component, and audio coding is carried out using one or more vectors of old edition audio coder to rearranged sequence to generate rearranged sequence One or more vectors encoded version.

The various aspects of technology described in the present invention aloow audio coding apparatus 510C according to following clause And it operates.

Clause 133143-1A.A kind of device, such as audio coding apparatus 510C comprising: one or more processors, institute It states one or more processors and is configured between one or more primary vectors and one or more secondary vectors and execute energy comparison With one or more primary vectors of the rearranged sequence of determination and promote one or more described primary vectors and it is described one or more second One or both extraction in vector, wherein one or more described primary vectors describe the sound field in the first part of audio data Distinct components, and one or more described secondary vectors describe the distinct components of the sound field in the second part of audio data.

Clause 133143-2A.The device of clause 133143-1A, wherein one or more described primary vectors do not indicate audio The background component of sound field in the first part of data, and wherein one or more described secondary vectors do not indicate the of audio data The background component of sound field in two parts.

Clause 133143-3A.The device of clause 133143-1A, wherein one or more described processors are further configured To be executed between one or more described primary vectors and one or more described secondary vectors after executing the energy comparison Crosscorrelation is to identify one or more described primary vectors relevant to one or more described secondary vectors.

Clause 133143-4A.The device of clause 133143-1A, wherein one or more described processors are further configured To perform the following operation: it is more one or more than described to generate to give up one or more of described secondary vector based on the energy comparison A secondary vector is reduced one or more secondary vectors with less vector;In one or more described primary vectors at least One and it is described reduced between one or more secondary vectors execute crosscorrelation with identify it is described reduced one or more second In vector in one or more described primary vectors described at least one relevant secondary vector;And it is based on the intersection phase It is reordered at least one of one or more described primary vectors to generate one or more primary vectors of rearranged sequence pass.

Clause 133143-5A.The device of clause 133143-1A, wherein one or more described processors are further configured To perform the following operation: one or more of secondary vector is given up based on the energy comparison with generate than it is described one or more the Two vectors are reduced one or more secondary vectors with less vector;In at least one of one or more described primary vectors With through reduce between one or more secondary vectors execute crosscorrelation with identify through reduce in one or more secondary vectors with institute State the relevant secondary vector of at least one of one or more primary vectors;Based on the crosscorrelation to it is described one or more At least one of one vector reorders to generate one or more primary vectors of rearranged sequence;And to the one or more of rearranged sequence A primary vector is encoded to generate the audio coding version of one or more primary vectors of rearranged sequence.

Clause 133143-6A.The device of clause 133143-1A, wherein one or more described processors are further configured To perform the following operation: one or more of secondary vector is given up based on the energy comparison with generate than it is described one or more the Two vectors are reduced one or more secondary vectors with less vector;In at least one of one or more described primary vectors With through reduce between one or more secondary vectors execute crosscorrelation with identify through reduce in one or more secondary vectors with institute State the relevant secondary vector of at least one of one or more primary vectors;Based on the crosscorrelation to one or more At least one of one vector reorders to generate one or more primary vectors of rearranged sequence;To one or more of rearranged sequence Primary vector is encoded to generate the audio coding version of one or more primary vectors of rearranged sequence;And bit stream is generated to wrap The encoded version of one or more primary vectors containing rearranged sequence.

Clause 133143-7A.The device of technical solution 3A to 6A, wherein the first part of the audio data includes having First audio frame of M sample, wherein the second part of the audio data includes second sound with M sample of same number Frequency frame, wherein one or more described processors be further configured to when executing crosscorrelation relative to one or more first to The last M-Z value of at least one of amount and the preceding M-Z value for being reduced each of one or more secondary vectors execute Crosscorrelation with identify through reduce in one or more secondary vectors in one or more primary vectors described at least one phase The secondary vector closed, and wherein Z is less than M.

Clause 133143-8A.The device of technical solution 3A to 6A, wherein the first part of the audio data includes having First audio frame of M sample, wherein the second part of the audio data includes second sound with M sample of same number Frequency frame, wherein one or more described processors be further configured to when executing crosscorrelation relative to one or more first to The last M-Y value of at least one of amount and the preceding M-Z value for being reduced each of one or more secondary vectors execute Crosscorrelation with identify through reduce in one or more secondary vectors in one or more primary vectors described at least one phase The secondary vector closed, and wherein both Z and Y are less than M.

Clause 133143-9A.The device of technical solution 3A to 6A, wherein one or more described processors are further configured Invert at least one of one or more primary vectors and one or more secondary vectors when executing crosscorrelation.

Clause 133143-10A.The device of clause 133143-1A, wherein one or more described processors are further configured With relative to indicate sound field multiple spherical harmonic coefficients execute singular value decomposition with generate one or more primary vectors and one or more Secondary vector.

Clause 133143-11A.The device of clause 133143-1A, wherein one or more described processors are further configured With relative to indicate sound field multiple spherical harmonic coefficients execute singular value decomposition to generate the left surprise for indicating the multiple spherical harmonic coefficient The U matrix of incorgruous amount, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and the right side for indicating the multiple spherical harmonic coefficient The Y matrix of singular vector, and one or more described primary vectors and one or more described secondary vectors are generated as the U square The function of one or more of battle array, s-matrix and V matrix.

Clause 133143-12A.The device of clause 133143-1A, wherein one or more described processors are further configured Indicate that a left side for the multiple spherical harmonic coefficient is unusual to generate relative to multiple spherical harmonic coefficient row singular value decompositions of sound field are indicated The right side of the U matrix of vector, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and the multiple spherical harmonic coefficient of expression is unusual The V matrix of vector executes high-lighting analysis relative to the s-matrix to identify one or more U of U matrixDISTVector and s-matrix One or more SDISTVector, and by least partly by one or more described UDISTVector is multiplied by one or more described SDIST Vector and determine one or more described primary vectors and one or more described secondary vectors.

Clause 133143-13A.The device of clause 133143-1A, wherein the first part of the audio data is in time Before the second part for coming across the audio data.

Clause 133143-14A.The device of clause 133143-1A, wherein the first part of the audio data is in time After the second part for coming across the audio data.

Clause 133143-15A.The device of clause 133143-1A, wherein one or more described processors are further configured To calculate the root mean square energy of each of one or more primary vectors and one or more secondary vectors when executing energy comparison Amount, and compare the root mean square energy calculated at least one of one or more described primary vectors be directed to it is described one or more The root mean square energy that each of a secondary vector calculates.

Clause 133143-16A.The device of clause 133143-1A, wherein one or more described processors are further configured Reordered at least one of one or more primary vectors based on the energy comparison to generate the one or more of rearranged sequence A primary vector, and wherein one or more described processors are further configured to when reordering to primary vector described one Or apply fade in/out interpolation window to ensure when generating one or more primary vectors of rearranged sequence between multiple primary vectors Smooth transition.

Clause 133143-17A.The device of clause 133143-1A, wherein one or more described processors are further configured With at least reordered based on the energy comparison to one or more primary vectors with generate rearranged sequence one or more first to Amount generates bit stream with the warp knit of one or more primary vectors comprising rearranged sequence or one or more primary vectors of rearranged sequence Code version, and the information that reorders for specifying description how to reorder to one or more described primary vectors in the bit stream.

Clause 133143-18A.The device of clause 133143-1A, wherein the energy comparison promote one or more first to Amount and one or more secondary vectors in one or both extraction so as to facilitate one or more primary vectors and one or more second Described one or both audio coding in vector.

Clause 133143-1B.A kind of device, such as audio coding apparatus 510C comprising: one or more processors, institute One or more processors are stated to be configured to execute crosscorrelation relative to one or more primary vectors and one or more secondary vectors With one or more primary vectors of the rearranged sequence of determination and promote one or more described primary vectors and it is described one or more second One or both extraction in vector, wherein one or more described primary vectors describe the sound field in the first part of audio data Distinct components, and one or more described secondary vectors describe the distinct components of the sound field in the second part of audio data.

Clause 133143-2B.The device of clause 133143-1B, wherein one or more described primary vectors do not indicate audio The background component of sound field in the first part of data, and wherein one or more described secondary vectors do not indicate the of audio data The background component of sound field in two parts.

Clause 133143-3B.The device of clause 133143-1B, wherein one or more described processors are further configured To execute energy comparison between one or more primary vectors and one or more secondary vectors before executing the crosscorrelation Reduced one or more secondary vectors with less vector than one or more described secondary vectors to generate, and wherein described one Or multiple processors be further configured to when executing the crosscorrelation one or more primary vectors with through reduction one or Crosscorrelation is executed between multiple secondary vectors to promote to one in one or more primary vectors and one or more secondary vectors Or both audio coding.

Clause 133143-4B.The device of clause 133143-3B, wherein one or more described processors are further configured To calculate the root mean square energy of each of one or more primary vectors and one or more secondary vectors when executing energy comparison Amount, and compare the root mean square energy calculated at least one of one or more described primary vectors be directed to it is described one or more The root mean square energy that each of a secondary vector calculates.

Clause 133143-5B.The device of clause 133143-3B, wherein one or more described processors are further configured To perform the following operation: one or more of secondary vector is given up based on the energy comparison with generate than it is described one or more the Two vectors are reduced one or more secondary vectors with less vector;Wherein one or more described processors are further configured With when executing the crosscorrelation at least one of one or more described primary vectors with reduced one or more second Between vector execute crosscorrelation with identify through reduce in one or more secondary vectors in one or more described primary vectors At least one relevant secondary vector;And wherein one or more described processors are further configured to based on the friendship Fork correlation reorder at least one of one or more described primary vectors with generate rearranged sequence one or more first to Amount.

Clause 133143-6B.The device of clause 133143-3B, wherein one or more described processors are further configured With based on the energy comparison give up one or more of secondary vector with generate than one or more secondary vectors with it is less to Amount is reduced one or more secondary vectors, wherein one or more described processors are further configured to execution crosscorrelation When at least one of one or more primary vectors with reduced between one or more secondary vectors execute crosscorrelation to know Not through reducing the secondary vector relevant at least one of one or more primary vectors in one or more secondary vectors, And wherein one or more described processors are further configured to based on the crosscorrelation in one or more primary vectors At least one reorders to generate one or more primary vectors of rearranged sequence, and to one or more primary vectors of rearranged sequence It is encoded to generate the audio coding version of one or more primary vectors of rearranged sequence.

Clause 133143-7B.The device of clause 133143-3B, wherein one or more described processors are further configured With based on the energy comparison give up one or more of secondary vector with generate than one or more secondary vectors with it is less to Amount is reduced one or more secondary vectors, wherein one or more described processors are further configured to execution crosscorrelation When at least one of one or more primary vectors with reduced between one or more secondary vectors execute crosscorrelation to know Not through reducing the secondary vector relevant at least one of one or more primary vectors in one or more secondary vectors, And wherein one or more described processors are further configured to based on the crosscorrelation in one or more primary vectors At least one reorders to generate one or more primary vectors of rearranged sequence, to one or more primary vectors of rearranged sequence into Row coding generates bit stream to generate the audio coding version of one or more primary vectors of rearranged sequence to include rearranged sequence One or more primary vectors encoded version.

Clause 133143-8B.The device of technical solution 3B to 7B, wherein the first part of the audio data includes having First audio frame of M sample, wherein the second part of the audio data includes second sound with M sample of same number Frequency frame, wherein one or more described processors be further configured to when executing crosscorrelation relative to one or more first to The last M-Z value of at least one of amount and the preceding M-Z value for being reduced each of one or more secondary vectors execute Crosscorrelation with identify through reduce in one or more secondary vectors in one or more primary vectors described at least one phase The secondary vector closed, and wherein Z is less than M.

Clause 133143-9B.The device of technical solution 3B to 7B, wherein the first part of the audio data includes having First audio frame of M sample, wherein the second part of the audio data includes second sound with M sample of same number Frequency frame, wherein one or more described processors be further configured to when executing crosscorrelation relative to one or more first to The last M-Y value of at least one of amount and the preceding M-Z value for being reduced each of one or more secondary vectors execute Crosscorrelation with identify through reduce in one or more secondary vectors in one or more primary vectors described at least one phase The secondary vector closed, and wherein both Z and Y are less than M.

Clause 133143-10B.The device of technical solution 1B, wherein one or more described processors are further configured to Invert at least one of one or more primary vectors and one or more secondary vectors when executing crosscorrelation.

Clause 133143-11B.The device of clause 133143-1B, wherein one or more described processors are further configured With relative to indicate sound field multiple spherical harmonic coefficients execute singular value decomposition with generate one or more primary vectors and one or more Secondary vector.

Clause 133143-12B.The device of clause 133143-1B, wherein one or more described processors are further configured With relative to indicate sound field multiple spherical harmonic coefficients execute singular value decomposition to generate the left surprise for indicating the multiple spherical harmonic coefficient The U matrix of incorgruous amount, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and the right side for indicating the multiple spherical harmonic coefficient The V matrix of singular vector, and one or more described primary vectors and one or more described secondary vectors are generated as the U square The function of one or more of battle array, s-matrix and V matrix.

Clause 133143-13B.The device of clause 133143-1B, wherein one or more described processors are further configured Indicate that a left side for the multiple spherical harmonic coefficient is unusual to generate relative to multiple spherical harmonic coefficient row singular value decompositions of sound field are indicated The right side of the U matrix of vector, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and the multiple spherical harmonic coefficient of expression is unusual The V matrix of vector executes high-lighting analysis relative to the s-matrix to identify one or more U of U matrixDISTVector and s-matrix One or more SDISTVector, and by least partly by one or more described UDISTVector is multiplied by one or more described SDIST Vector and determine one or more described primary vectors and one or more described secondary vectors.

Clause 133143-14B.The device of clause 133143-1B, wherein one or more described processors are further configured With relative to indicate sound field multiple spherical harmonic coefficients execute singular value decomposition to generate the left surprise for indicating the multiple spherical harmonic coefficient The right singular vector of the U matrix of incorgruous amount, the s-matrix of the singular value of expression spherical harmonic coefficient and the multiple spherical harmonic coefficient of expression V matrix, and high-lighting point is executed relative to the s-matrix when determining one or more primary vectors and one or more secondary vectors Analysis is with by one or more V of V matrixDISTVector is identified as one or more described primary vectors and one or more described secondary vectors At least one of.

Clause 133143-15B.The device of clause 133143-1B, wherein the first part of the audio data is in time Before the second part for coming across the audio data.

Clause 133143-16B.The device of clause 133143-1B, wherein the first part of the audio data is in time After the second part for coming across the audio data.

Clause 133143-17B.The device of clause 133143-1B, wherein one or more described processors are further configured Reordered at least one of one or more primary vectors based on the crosscorrelation to generate the one or more of rearranged sequence A primary vector, and when reordering to primary vector between one or more primary vectors using fade in/out interpolation window with Ensure the smooth transition when generating one or more primary vectors of rearranged sequence.

Clause 133143-18B.The device of clause 133143-1B, wherein one or more described processors are further configured With at least reordered based on the crosscorrelation to one or more primary vectors with generate rearranged sequence one or more first to Amount generates bit stream with the warp knit of one or more primary vectors comprising rearranged sequence or one or more primary vectors of rearranged sequence Code version, and specify how to reorder to one or more described primary vectors in the bit stream.

Clause 133143-19B.The device of clause 133143-1B, wherein the crosscorrelation promote one or more first to Amount and one or more secondary vectors in one or both extraction so as to facilitate one or more primary vectors and one or more second Described one or both audio coding in vector.

Figure 40 D is that the various aspects of technology described in the present invention can be performed with packed description two dimension or three dimensional sound in explanation The block diagram of the example audio code device 510D of the spherical harmonic coefficient of field.Audio coding apparatus 510D can be similar to audio coding apparatus 510C, similar place are that audio coding apparatus 510D includes that audio compressing unit 512, audio coding unit 514 and bit stream produce Raw unit 516.In addition, the audio compressing unit 512 of audio coding apparatus 510D can be similar to the sound of audio coding apparatus 510C Frequency compression unit, similar place are that audio compressing unit 512 includes decomposition unit 518.

However, the audio compressing unit 512 of audio coding apparatus 510D may differ from the audio of audio coding apparatus 510C Compression unit 512, the difference is that sound field component extraction unit 520 includes to be expressed as 534 (" quantifying unit of quantifying unit 534 ") additional unit.For this reason, the sound field component extraction unit 520 of audio coding apparatus 510D is expressed as " sound field point Measure extraction unit 520D ".

The expression of quantifying unit 534 is configured to quantify one or more VT DISTVector 525E and/or one or more VT BGVector 525F is to generate one or more corresponding VT Q_DISTVector 525G and/or one or more VT Q_BGThe unit of vector 525H.Quantization is single Member 534 can quantify (it is for by eliminating the signal processing term being rounded to indicate the mathematics that the position of value carries out) one or Multiple VT DISTVector 525E is to reduce to indicate one or more V in bit stream 517T DISTThe number of the position of vector 525E.? In some examples, quantifying unit 534 can quantify one or more VT DIST32 place values of vector 525E replace this with 16 place values are rounded A little 32 place values are to generate one or more VT Q_DISTVector 525G.In in this respect, quantifying unit 534 can be similar to opposite above Mode described in the quantifying unit 52 of the audio coding apparatus 20 shown in the example of Fig. 4 operates.

The quantization of this property can introduce error into the expression of sound field, and the error becomes according to the roughness of quantization Change.In other words, to indicate one or more VT DISTThe position of vector 525E is more, and the quantization error that may cause is fewer.It can lead to It crosses from one or more VT Q_DISTVector 525G subtracts one or more VT DISTVector 525E and determine be attributed to VT DISTVector 525E's Quantization and caused by quantization error (it is represented by " EDIST”)。

Technology described according to the present invention, audio coding apparatus 510D can be by by EDISTError pro to pass through by One or more UBGVector 525D is multiplied by one or more SBGVector 525B and then multiplied by one or more VT BGVector 525F and generate UDIST*SDISTOne or more of vector 527 or background spherical harmonic coefficient 531 modify the U with other other waysDIST* SDISTOne or more of vector 527 or background spherical harmonic coefficient 531 and compensate EDISTOne or more of quantization error.Some In example, audio coding apparatus 510D can compensate only for UDIST*SDISTE in vector 527DISTError.In other examples, audio Code device 510D can compensate only for the E in background spherical harmonic coefficientBGError.In other examples again, audio coding apparatus 510D can Compensate UDIST*SDISTE in vector 527 and background spherical harmonic coefficient the twoDISTError.

In operation, prominent Component Analysis unit 524 can be configured with by one or more SDISTVector 525, one or more SBGVector 525B, one or more UDISTVector 525C, one or more UBGVector 525D, one or more VT DISTVector 525E and one Or multiple VT BGVector 525F is output to mathematical units 526.Prominent Component Analysis unit 524 can also be by one or more VT DISTVector 525E is output to quantifying unit 534.Quantifying unit 534 can quantify one or more VT DISTVector 525E is to generate one or more VT Q_DISTVector 525G.Quantifying unit 534 can be by one or more VT Q_DISTVector 525G is provided to mathematical units 526, while also will One or more VT Q_DISTVector 525G is provided to vector and is reordered unit 532 (as described above).Vector reorders unit 532 It can be similar to above in relation to VT DISTMode described in vector 525E is relative to one or more described VT Q_DISTVector 525G Operation.

After receiving these vectors 525 to 525G (" vector 525 "), mathematical units 526 can determination be retouched first at once It states the different spherical harmonic coefficient of the distinct components of sound field and describes the background spherical harmonic coefficient of the background component of sound field.Matrix mathematical units 526 can be configured by by one or more UDIST525C vector is multiplied by one or more SDISTVector 525A and then multiplied by one or Multiple VT DISTVector 525E and determine different spherical harmonic coefficient.Mathematical units 526 can be configured by by one or more UBG 525D vector is multiplied by one or more SBGVector 525A and then multiplied by one or more VT BGVector 525E and determine the humorous system of background ball Number.

Mathematical units 526 can be then by relative to one or more VT Q_DISTVector 525G executes pseudo- inverse operation and then will Different ball is humorous multiplied by one or more VT Q_DISTThe pseudoinverse of vector 525G and determine one or more compensated UDIST*SDISTVector 527'(its can be similar to UDIST*SDISTVector 527, but these vectors include to compensate EDISTThe value of error).Vector reorders Unit 532 can be operated in a manner of as described above to generate the vector 527' of rearranged sequence, the vector of the rearranged sequence 527' then carries out audio coding to generate the vector 515' of the rearranged sequence of audio coding, again such as by audio coding unit 515A It is described above.

Mathematical units 526 can be next by EDISTError pro is to background spherical harmonic coefficient.To execute this projection, mathematical units 526 can determine or restore in other ways original spherical harmonic coefficient and different spherical harmonic coefficient is added to background spherical harmonic coefficient 511.Mathematical units 526 then can subtract quantified different spherical harmonic coefficient from spherical harmonic coefficient 511, and (it can be by by UDISTVector 525C is multiplied by SDISTVector 525A and then multiplied by VT Q_DISTVector 525G and generate) and background spherical harmonic coefficient with determine be attributed to VT DISTThe quantization of vector 519 and caused by residual error.This error then can be added to quantified background ball by mathematical units 526 Humorous coefficient is to generate compensated quantified background spherical harmonic coefficient 531'.

Under any circumstance, Degree reduction unit 528A can be executed as described above with by compensated quantified back Scape spherical harmonic coefficient 531' is reduced to be reduced background spherical harmonic coefficient 529', can be by audio coding through reduction background spherical harmonic coefficient 529' The mode described above of unit 514 carry out audio coding with generate audio coding through reduce background spherical harmonic coefficient 515B'.

By this method, the technology aloows one or more components of audio coding apparatus 510D quantization means sound field One or more primary vectors (such as VT DISTVector 525E), and compensate the quantization for being attributed to one or more primary vectors and introduce In one or more secondary vectors (such as the U for one or more the identical components for also illustrating that sound fieldDIST*SDISTVector 527 and/or back The vector of scape spherical harmonic coefficient 531) in error.

In addition, the technology can provide this quantization error compensation according to following clause.

Clause 133146-1B.A kind of device, such as audio coding apparatus 510D comprising: one or more processors, institute One or more primary vectors that one or more processors are configured to one or more distinct components of quantization means sound field are stated, and are mended It repays the quantization for being attributed to one or more primary vectors and is incorporated in the one of one or more the identical distinct components for also illustrating that sound field Or the error in multiple secondary vectors.

Clause 133146-2B.The device of clause 133146-1B, wherein one or more described processors are configured to quantify From turning at least partially through the V matrix that multiple spherical harmonic coefficients relative to description sound field execute singular value decomposition and generate One or more vectors set.

Clause 133146-3B.The device of clause 133146-1B, wherein one or more described processors are further configured With relative to indicate sound field multiple spherical harmonic coefficients execute singular value decomposition to generate the left surprise for indicating the multiple spherical harmonic coefficient The U matrix of incorgruous amount, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and the right surprise for indicating the multiple spherical harmonic coefficient The V matrix of incorgruous amount, and wherein one or more described processors be configured to quantify the transposition from V matrix one or more to Amount.

Clause 133146-4B.The device of clause 133146-1B, wherein one or more described processors are configured to relatively Singular value decomposition is executed in multiple spherical harmonic coefficients of expression sound field to generate the left singular vector for indicating the multiple spherical harmonic coefficient U matrix, indicate the multiple spherical harmonic coefficient singular value s-matrix and indicate the right singular vector of the multiple spherical harmonic coefficient V matrix, wherein one or more described processors are configured to quantify one or more vectors of the transposition from V matrix, and its Described in one or more processors be configured to compensate for being attributed to quantization and being incorporated in by by one or more U vectors of U matrix The error in one or more calculated U*S vectors multiplied by one or more S vectors of s-matrix.

Clause 133146-5B.The device of clause 133146-1B, wherein one or more described processors are further configured To perform the following operation: indicating the multiple ball relative to the multiple spherical harmonic coefficients execution singular value decomposition for indicating sound field to generate The U matrix of the left singular vector of humorous coefficient, indicate the multiple spherical harmonic coefficient singular value s-matrix and indicate the multiple ball The V matrix of the right singular vector of humorous coefficient;Determine one or more U of U matrixDISTVector, each of these person correspond to sound field One of distinct components;Determine one or more S of s-matrixDISTVector, each of these person correspond to different point of sound field Identical one in amount;And determine one or more V of the transposition of V matrixT DISTVector, each of these person correspond to the phase of sound field Identical one in different component.

Wherein one or more described processors are configured to quantify one or more VT DISTVector is to generate one or more VT Q_DISTVector, and wherein one or more described processors are configured to compensate for being attributed to quantization and be incorporated in by by U matrix One or more UDISTVector multiplied by s-matrix one or more SDISTVector and one or more calculated UDIST*SDISTIn vector Error to generate one or more U through error compensationDIST*SDISTVector.

Clause 133146-6B.The device of clause 133146-5B, wherein one or more described processors are configured to be based on One or more described UDISTVector, one or more described SDISTVector and one or more described VT DISTVector determines the different humorous system of ball Number, and relative to the VT Q_DISTVector execute pseudoinverse with by the different spherical harmonic coefficient divided by one or more described VT Q_DISTTo It measures and generates at least partly compensation in turn and pass through VT DISTThe quantization of vector and the error that introduces through the one or more of error compensation A UC_DIST*SC_DISTVector.

Clause 133146-7B.The device of clause 133146-5B, wherein one or more described processors are further configured To one or more U through error compensationDIST*SDISTVector carries out audio coding.

Clause 133146-8B.The device of clause 133146-1B, wherein one or more described processors are further configured With relative to indicate sound field multiple spherical harmonic coefficients execute singular value decomposition to generate the left surprise for indicating the multiple spherical harmonic coefficient The U matrix of incorgruous amount, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and the right side for indicating the multiple spherical harmonic coefficient The V matrix of singular vector determines one or more U of one or more background components of the description sound field of the U matrixBGVector and One or more U of one or more distinct components of the description sound field of the U matrixDISTVector determines the description sound of the s-matrix One or more S of one or more background components of fieldBGOne or more distinct components of the description sound field of vector and the s-matrix One or more SDISTVector, and determine one or more V of the transposition of the V matrixT DISTOne or more V of vector sumT BGVector, The wherein VT DISTOne or more distinct components and the V of vector description sound fieldT BGOne or more backgrounds point of sound field are described Amount, wherein one or more described processors are configured to quantify one or more described VT DISTVector is to generate one or more VT Q_DIST Vector, and wherein one or more described processors are further configured to compensate for being attributed to quantization and be incorporated in by will be one or more A UBGVector is multiplied by one or more SBGVector and then multiplied by one or more VT BGVector and in the background spherical harmonic coefficient that is formed At least part of error is to generate the background spherical harmonic coefficient through error compensation.

Clause 133146-9B.The device of clause 133146-8B, wherein one or more described processors are configured to be based on VT DISTVector and by by the UDISTVector is multiplied by SDISTVector and one or more U formedDIST*SDISTDescribed in vector determines Error, and the identified error is added to background spherical harmonic coefficient to generate the background spherical harmonic coefficient through error compensation.

Clause 133146-10B.The device of clause 133146-8B, wherein one or more described processors are further configured To carry out audio coding to the background spherical harmonic coefficient through error compensation.

Clause 133146-11B.The device of clause 133146-1B,

Wherein one or more described processors are configured to compensate for being attributed to the quantization of one or more primary vectors and introducing Error in one or more secondary vectors of one or more identical components for also illustrating that sound field is to generate one or more through accidentally The secondary vector of difference compensation, and wherein one or more described processors are further configured to generate bit stream comprising one or more Secondary vector and one or more quantified primary vectors through error compensation.

Clause 133146-12B.The device of clause 133146-1B, wherein one or more described processors are configured to compensate for Due to the quantization of one or more primary vectors and be incorporated in one or more for also illustrating that one or more identical components of sound field Error in secondary vector to generate one or more secondary vectors through error compensation, and wherein one or more described processors into One step is configured to the secondary vector to one or more through error compensation and carries out audio coding, and generates bit stream to compile comprising audio One or more secondary vectors and one or more quantified primary vectors through error compensation of code.

Clause 133146-1C.A kind of device, such as audio coding apparatus 510D comprising: one or more processors, institute One or more primary vectors that one or more processors are configured to one or more distinct components of quantization means sound field are stated, and are mended Repay the quantization for being attributed to one or more primary vectors and be incorporated in one or more background components for indicating sound field one or more the Error in two vectors.

Clause 133146-2C.The device of clause 133146-1C, wherein one or more described processors are configured to quantify From turning at least partially through the V matrix that multiple spherical harmonic coefficients relative to description sound field execute singular value decomposition and generate One or more vectors set.

Clause 133146-3C.The device of clause 133146-1C, wherein one or more described processors are further configured With relative to indicate sound field multiple spherical harmonic coefficients execute singular value decomposition to generate the left surprise for indicating the multiple spherical harmonic coefficient The U matrix of incorgruous amount, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and the right surprise for indicating the multiple spherical harmonic coefficient The V matrix of incorgruous amount, and wherein one or more described processors be configured to quantify the transposition from V matrix one or more to Amount.

Clause 133146-4C.The device of clause 133146-1C, wherein one or more described processors are further configured With relative to indicate sound field multiple spherical harmonic coefficients execute singular value decomposition to generate the left surprise for indicating the multiple spherical harmonic coefficient The U matrix of incorgruous amount, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and the multiple spherical harmonic coefficient of expression are unusual The V matrix of vector right singular vector;Determine one or more U of the U matrixDISTVector, each of these person correspond to sound field One of distinct components;Determine one or more vectors of the s-matrix, each of these person corresponds to different point of sound field Same one in amount;And determine one or more V of the transposition of V matrixT DISTVector, each of these person correspond to the phase of sound field Same one in different component, wherein one or more described processors are configured to quantify one or more described VT DISTVector is to produce One or more raw VT Q_DISTVector, and compensate and be attributed to quantization and be incorporated in by by one or more U of U matrixDISTVector multiplied by One or more S of s-matrixDISTVector and one or more calculated UDIST*SDISTError in vector is to generate one or more U through error compensationDIST*SDISTVector.

Clause 133146-5C.The device of clause 133146-4C, wherein one or more described processors are configured to be based on One or more described UDISTVector, one or more described SDISTVector and one or more described VT DISTVector determines the different humorous system of ball Number, and relative to the VT Q_DISTVector execute pseudoinverse with by the different spherical harmonic coefficient divided by one or more described VT Q_DISTTo It measures and generates at least partly compensation in turn and pass through VT DISTThe quantization of vector and one or more U of error introducedC_DIST*SC_DIST Vector.

Clause 133146-6C.The device of clause 133146-4C, wherein one or more described processors are further configured To one or more U through error compensationDIST*SDISTVector carries out audio coding.

Clause 133146-7C.The device of clause 133146-1C, wherein one or more described processors are further configured To perform the following operation: indicating the multiple ball relative to the multiple spherical harmonic coefficients execution singular value decomposition for indicating sound field to generate The U matrix of the left singular vector of humorous coefficient, indicate the multiple spherical harmonic coefficient singular value s-matrix and indicate the multiple ball The V matrix of the right singular vector of humorous coefficient;Determine one or more U of one or more background components of the description sound field of U matrixBGTo One or more U of one or more distinct components of amount and the description sound field of U matrixDISTVector;Determine the description sound field of s-matrix One or more S of one or more background componentsBGOne or more of one or more distinct components of the description sound field of vector and s-matrix SDISTVector;And determine one or more V of the transposition of V matrixT DISTVector and one or more VT BGVector, wherein the VT DISTTo One or more described distinct components and the V of amount description sound fieldT BGOne or more described background components of sound field are described, wherein One or more described processors are configured to quantify one or more described VT DISTVector is to generate one or more VT Q_DISTVector, and Wherein one or more described processors are configured to compensate for being attributed to quantization and being incorporated in through one or more U by described inBGVector Multiplied by one or more described SBGVector and then multiplied by one or more described VT BGVector and the mistake in the background spherical harmonic coefficient that is formed Difference is to generate the background spherical harmonic coefficient through error compensation.

Clause 133146-8C.The device of clause 133146-7C, wherein one or more described processors are configured to be based on VT DISTVector and by by the UDISTVector is multiplied by SDISTVector and one or more U formedDIST*SDISTDescribed in vector determines Error, and the identified error is added to background spherical harmonic coefficient to generate the background spherical harmonic coefficient through error compensation.

Clause 133146-9C.The device of clause 133146-7C, wherein one or more described processors are further configured To carry out audio coding to the background spherical harmonic coefficient through error compensation.

Clause 133146-10C.The device of clause 133146-1C, wherein one or more described processors are configured to compensate for Due to the quantization of one or more primary vectors and be incorporated in one or more for also illustrating that one or more identical components of sound field Error in secondary vector generates bit stream comprising one or more warps to generate one or more secondary vectors through error compensation The secondary vector of error compensation and one or more quantified primary vectors.

Clause 133146-11C.The device of clause 133146-1C, wherein one or more described processors are further configured The quantization of one or more primary vectors is attributed to compensation and is incorporated in one or more the identical components for also illustrating that sound field One or more secondary vectors in error to generate one or more secondary vectors through error compensation, to it is described one or more warp The secondary vector of error compensation carries out audio coding, and generate bit stream with comprising one or more through audio coding through error compensation Secondary vector and described one or more quantified primary vectors.

In other words, when the purpose reduced for bandwidth uses SVD (or correlation technique, example based on frame to HoA signal Such as KLT&PCA) decompose when, technology described in the present invention can enable audio coding apparatus 10D quantify the preceding several of U matrix The correspondence vector of vector (multiplied by the correspondence singular value of s-matrix) and V vector.This by include sound field ' prospect ' or ' different ' point Amount.The technology can then make audio coding apparatus 510D be able to use ' black box ' the audio coding engine such as AAC encoder U*S vector is decoded.V vector can scalar-quantized or vector quantization.

In addition, some in residual vector in U matrix can use multiplied by s-matrix and the correspondence singular value of V matrix and also ' black box ' audio coding engine is decoded.These are by ' background ' component including sound field.Simple 16 scalar quantizations of V vector About 80kbps overhead can lead to for quadravalence (25 coefficients) and 160kbps can lead to for 6 ranks (49 coefficients).It is thicker Quantization slightly can lead to bigger quantization error.Technology described in the present invention can be by ' throwing the quantization error of V vector Shadow ' to compensating the quantization error of V vector in prospect and background component.

Technology in the present invention may include calculating the quantified version of practical V vector.This quantified V vector can be described as V' (wherein V'=V+e).The basic HoA signal for prospect component that the technology attempts to rebuild is provided by H_f=USV, Middle U, S and V only contain foreground elements.For purposes of this discussion, US will be replaced by the single set of vector U.Thus, H_f= UV.It is assumed that we have the V' of error, then the technology attempts to rebuild H_f as closely as possible.Thus, the technology can make Audio coding apparatus 10D can find out the U' so that H_f=U'V'.Audio coding apparatus 10D, which can be used, allows U'=H_f [V'] The pseudoinverse technique of ^ (- 1).U' is decoded using so-called ' black box ' audio coding engine, the technology can minimize by In the V' vector for being referred to alternatively as error and caused by error in H.

In a similar manner, the technology can also enable audio coding apparatus will be due to quantization V and caused by error throw Shadow is into background element.Audio coding apparatus 510D can be configured to rebuild total HoA signal, be foreground and background HoA signal Combination, that is, H=H_f+H_b.Due to the quantization error in V', this can be modeled as H=H_f+e+H_b again.By this method, Instead of making H_b by ' black box tone decoder ', we make (e+H_b) by tone decoder, to essentially compensate for V' In error.In fact, this compensation is only up to determined the mistake for the order for thinking that background element is sent by audio coding apparatus 510D Difference.

Figure 40 E is that the various aspects of technology described in the present invention can be performed with packed description two dimension or three dimensional sound in explanation The block diagram of the example audio code device 510E of the spherical harmonic coefficient of field.Audio coding apparatus 510E can be similar to audio coding apparatus 510D, similar place are that audio coding apparatus 510E includes that audio compressing unit 512, audio coding unit 514 and bit stream produce Raw unit 516.In addition, the audio compressing unit 512 of audio coding apparatus 510E can be similar to the sound of audio coding apparatus 510D Frequency compression unit, similar place are that audio compressing unit 512 includes decomposition unit 518.

However, the audio compressing unit 512 of audio coding apparatus 510E may differ from the audio of audio coding apparatus 510D Compression unit 512, the difference is that the mathematical units 526 of sound field component extraction unit 520 execute it is described in the present invention The additional aspect of technology is to reduce V through reduction version taking a step forward included in bit stream 517 for the transposition of V matrix 519A Matrix 519A.For this reason, the sound field component extraction unit 520 of audio coding apparatus 510E is expressed as " sound field component extraction Unit 520E ".

In the example of Figure 40 E, it will not be reduced background spherical harmonic coefficient 529' and be forwarded to audio coding unit 514, rank Number reduction unit 528 will return to mathematical units 526 through reduction background spherical harmonic coefficient 529'.As noted above, these are through contracting Background subtraction spherical harmonic coefficient 529', which may pass through, removes corresponding to one or more identified orders and/or sub- rank in coefficient Those of several spherical substrate function coefficient and reduced.Through reduction background spherical harmonic coefficient 529' through reduction order can be by becoming Measure NBGIt indicates.

It is assumed that sound field component extraction unit 520E can not be relative to one or more U of rearranged sequenceDIST*SDISTVector 533' Execute Degree reduction, then describe the spherical harmonic coefficient of the distinct components of sound field this decompose order (it can be by variable NDISTIt indicates) Background order N can be greater thanBG.In other words, NBGN may be usually less thanDIST。NBGIt is likely less than NDISTA possible cause be false Background component is determined without many directionality, so that not needing higher-order spherical substrate function, and then is realized Degree reduction and is led Cause NBGLess than NDIST

It is assumed that previously sending one or more V of rearranged sequence publiclyT Q_DISTVector 539 is without by these 539 sounds of vector Frequency is encoded in bit stream 517, as shown in the example of Figure 40 A to 40D, then one or more V of rearranged sequenceT Q_DISTVector 539 Sizable bandwidth may be consumed.As an example, one or more V of rearranged sequenceT Q_DISTEach of vector 539 is worked as It is quantified to 16 scalar values when can disappear for fourth order ambiophony audio data (wherein each vector has 25 coefficients) Consume approximation 20Kbps and for the 6th rank ambiophony audio data (wherein each vector has 49 coefficients) consumption approximation 40Kbps。

The various aspects of technology described according to the present invention, sound field component extraction unit 520E can reduce needs and be directed to Spherical harmonic coefficient or its decompose (such as one or more V of rearranged sequenceT Q_DISTVector 539) specified position amount.In some examples In, mathematical units 526 can determine the V of rearranged sequence based on the reduced spherical harmonic coefficient 529' of orderT Q_DISTIt will be by vector 539 It removes and the V with those of order reduced spherical harmonic coefficient 529' recombination vector and rearranged sequenceT Q_DISTGeneral in vector 539 Form VT SMALLThose of vector 521 vector.That is, mathematical units 526 can determine the rank of the reduced spherical harmonic coefficient 529' of order Number, wherein this order is represented by NBG.The V of rearranged sequenceT Q_DISTVector 539 can have by variable NDISTThe order of expression, Middle NDISTGreater than order NBG

Mathematical units 526 can then dissect the V of rearranged sequenceT Q_DISTFirst N of vector 539BGOrder removes specified correspond to It is less than or equal to N in orderBGSpherical substrate function through decomposing those of spherical harmonic coefficient vector.These are removed rearranged The V of sequenceT Q_DISTVector 539 can then be used to form intermediate spherical harmonic coefficient, be by by the U of rearranged sequenceDIST*SDISTVector Expression in 533' corresponds to order and is less than or equal to NBGSpherical substrate function spherical harmonic coefficient 511 through decompose version that A little vectors multiplied by removed rearranged sequence VT Q_DISTVector 539 is to form intermediate different spherical harmonic coefficient.Mathematical units 526 can Modified background ball is generated and the different spherical harmonic coefficient in the centre is added to order reduced spherical harmonic coefficient 529' Humorous coefficient 537.This modified background spherical harmonic coefficient 537 then can be transmitted to audio coding unit 514 by mathematical units 526, Audio coding unit 514 carries out audio coding to these coefficients 537 to form the modified background spherical harmonic coefficient of audio coding 515B'。

Mathematical units 526 can then transmit one or more described VT SMALLVector 521 can indicate to indicate to correspond to order Greater than NBGAnd it is less than or equal to NDISTThose of the decomposed form of spherical harmonic coefficient 511 of spherical substrate function vector 539.At this In aspect, the coefficient of audio coding apparatus 20 shown in the executable example similar to Fig. 4 of mathematical units 526 reduces unit 46 Operation.Mathematical units 526 can be by one or more described VT SMALLVector 521 is transmitted to bitstream producing unit 516, and bit stream generates Unit 516 can produce bit stream 517 comprising VT SMALLVector 521 (usually original non-through audio coding form in its).It is assumed that VT SMALLVector 521 includes the V than rearranged sequenceT Q_DISTThe few vector of vector 539, then the technology can be by only in bit stream 517 In specify VT SMALLVector 521 and promote for less bits to be assigned to the V of rearranged sequenceT Q_DISTVector 539.

Although being shown as non-quantized, in some cases, audio coding apparatus 510E can quantify VT BGVector 525F. In some cases, such as when audio coding unit 514 is not to compressed background spherical harmonic coefficient, audio coding apparatus 510E can Quantify VT BGVector 525F.

By this method, the technology can make audio coding apparatus 510E can determine from will with background spherical harmonic coefficient recombinate with It reduces in one or more vectors of the spherical harmonic coefficient decomposition for the amount of position for needing to distribute to one or more vectors in bit stream extremely Few one, wherein the spherical harmonic coefficient describes sound field, and wherein the background spherical harmonic coefficient describes one or more of identical sound field Background component.

That is, the technology can enable audio coding apparatus 510E be configured in a manner of being indicated by following clause.

Clause 133149-1A.A kind of device, such as audio coding apparatus 510E comprising: one or more processors, institute State one or more processors be configured to determine from will with background spherical harmonic coefficient recombinate with reduce need one will be assigned in bit stream Or at least one of one or more vectors that the spherical harmonic coefficient of the amount of the position of multiple vectors decomposes, wherein the spherical harmonic coefficient is retouched Sound field is stated, and wherein background spherical harmonic coefficient describes one or more background components of identical sound field.

Clause 133149-2A.The device of clause 133149-1A, wherein one or more described processors are further configured With determined by removed in one or more described vectors from one or more described vectors at least one and generate described one Or multiple vectors through reduction gather.

Clause 133149-3A.The device of clause 133149-1A, wherein one or more described processors are further configured With that cross from one or more described vectors remove in one or more described vectors determined by least one and generate institute Gathering through reduction for one or more vectors is stated, at least one removed and background ball in one or more described vectors of recombination are humorous Coefficient to generate modified background spherical harmonic coefficient, and generate bit stream with comprising one or more vectors through reduction set and The modified background spherical harmonic coefficient.

Clause 133149-4A.The device of clause 133149-3A, wherein one or more vectors are wrapped through reduction set It is contained in the bit stream without first through audio coding.

Clause 133149-5A.The device of clause 133149-1A, wherein one or more described processors are further configured With determined by removed in one or more described vectors from one or more described vectors at least one and generate described one Or at least one removed and background spherical harmonic coefficient being reduced set, recombinate in one or more described vectors of multiple vectors To generate modified background spherical harmonic coefficient, audio coding is carried out to modified background spherical harmonic coefficient, and generate bit stream to wrap The modified background spherical harmonic coefficient through reduction set and the audio coding containing one or more vectors.

Clause 133149-6A.The device of clause 133149-1A, wherein one or more described vectors include indicating sound field The vector at least in a certain respect of one or more distinct components.

Clause 133149-7A.The device of clause 133149-1A, wherein one or more described vectors include from least portion The one of the transposition for the V matrix for dividing ground to generate and executing singular value decomposition relative to the multiple spherical harmonic coefficient of description sound field Or multiple vectors.

Clause 133149-8A.The device of clause 133149-1A, wherein one or more described processors are further configured To execute singular value decomposition relative to the multiple spherical harmonic coefficient to generate the left singular vector for indicating the multiple spherical harmonic coefficient U matrix, indicate the multiple spherical harmonic coefficient singular value s-matrix and indicate the right singular vector of the multiple spherical harmonic coefficient V matrix, and wherein one or more described vectors include one or more vectors of the transposition from the V matrix.

Clause 133149-9A.The device of clause 133149-1A, wherein one or more described processors are further configured It removes corresponding in background spherical harmonic coefficient to execute Degree reduction relative to background spherical harmonic coefficient and there is identified order And/or those of spherical substrate function of sub- order background spherical harmonic coefficient, wherein the background spherical harmonic coefficient corresponds to order NBG

Clause 133149-10A.The device of clause 133149-1A, wherein one or more described processors are further configured It removes corresponding in background spherical harmonic coefficient to execute Degree reduction relative to background spherical harmonic coefficient and there is identified order And/or those of spherical substrate function of sub- order background spherical harmonic coefficient, wherein the background spherical harmonic coefficient, which corresponds to, is less than phase The order N of different spherical harmonic coefficientDISTOrder NBG, and wherein the different spherical harmonic coefficient indicates the distinct components of sound field.

Clause 133149-11A.The device of clause 133149-1A, wherein one or more described processors are further configured It removes corresponding in background spherical harmonic coefficient to execute Degree reduction relative to background spherical harmonic coefficient and there is identified order And/or those of spherical substrate function of sub- order background spherical harmonic coefficient, wherein the background spherical harmonic coefficient, which corresponds to, is less than phase The order N of different spherical harmonic coefficientDISTOrder NBG, and wherein the different spherical harmonic coefficient indicates the distinct components of sound field and is not subjected to Degree reduction.

Clause 133149-12A.The device of clause 133149-1A, wherein one or more described processors are further configured To execute singular value decomposition relative to the multiple spherical harmonic coefficient to generate the left singular vector for indicating the multiple spherical harmonic coefficient U matrix, indicate the multiple spherical harmonic coefficient singular value s-matrix and indicate the right singular vector of the multiple spherical harmonic coefficient V matrix, and determine V matrix transposition one or more VT DISTVector and one or more VT BGVector, it is described one or more VT DISTOne or more distinct components of vector description sound field, and one or more described VT BGOne or more back of vector description sound field Scape component, and wherein one or more described vectors include one or more described VT DISTVector.

Clause 133149-13A.The device of clause 133149-1A, wherein one or more described processors are further configured To execute singular value decomposition relative to the multiple spherical harmonic coefficient to generate the left singular vector for indicating the multiple spherical harmonic coefficient U matrix, indicate the multiple spherical harmonic coefficient singular value s-matrix and indicate the right singular vector of the multiple spherical harmonic coefficient V matrix, determine one or more V of the transposition of V matrixT DISTVector and one or more VT BGVector, one or more described VDISTTo One or more distinct components of amount description sound field, and one or more described VBGOne or more background components of vector description sound field, And quantify one or more described VT DISTVector is to generate one or more VT Q_DISTVector, and wherein one or more described vectors include One or more described VT Q_DISTVector.

Clause 133149-14A.The device of any one of clause 133149-12A or clause 133149-13A, wherein institute It states one or more processors to be further configured to perform the following operation: determining one or more U of U matrixDISTVector and one or Multiple UBGVector, one or more described UDISTOne or more distinct components of vector description sound field, and one or more described UBGTo One or more background components of amount description sound field;And determine one or more S of s-matrixDISTVector and one or more SBGVector, institute State one or more SDISTOne or more distinct components of vector description sound field, and one or more described SBGThe one of vector description sound field Or multiple background components.

Clause 133149-15A.The device of clause 133149-14A, wherein one or more described processors are further through matching It sets to perform the following operation: according to one or more UBGVector, one or more SBGVector and one or more VT BGDetermine that background ball is humorous Coefficient;Degree reduction is executed relative to background spherical harmonic coefficient to generate order equal to NBGThrough reduce background spherical harmonic coefficient;By one Or multiple UDISTMultiplied by one or more SDISTVector is to generate one or more UDIST*SDISTVector;It is moved from one or more described vectors Except determined by one or more described vectors at least one to generate gathering through reduction for one or more vectors;By one Or multiple UDIST*SDISTVector is multiplied by one or more described VT DISTVector or one or more described VT Q_DISTBeing removed in vector At least one to generate intermediate different spherical harmonic coefficient;And intermediate different spherical harmonic coefficient is added to background spherical harmonic coefficient to recombinate One or more described VT DISTVector or one or more described VT Q_DISTAt least one removed and the humorous system of background ball in vector Number.

Clause 133149-16A.The device of clause 133149-14A, wherein one or more described processors are further through matching It sets to perform the following operation: according to one or more UBGVector, one or more SBGVector and one or more VT BGDetermine that background ball is humorous Coefficient;Degree reduction is executed relative to background spherical harmonic coefficient to generate order equal to NBGThrough reduce background spherical harmonic coefficient;By one Or multiple UDISTMultiplied by one or more SDISTVector is to generate one or more UDIST*SDISTVector;Reorder it is described one or more UDIST*SDISTVector is to generate one or more U of rearranged sequenceDIST*SDISTVector;Described in the removal of one or more described vectors At least one identified gathering through reduction with generation one or more vectors in one or more vectors;By rearranged sequence One or more UDIST*SDISTVector is multiplied by one or more described VT DISTVector or one or more described VT Q_DISTInstitute in vector At least one removed is to generate intermediate different spherical harmonic coefficient;And by intermediate different spherical harmonic coefficient be added to background spherical harmonic coefficient with Recombinate one or more described VT DISTVector or one or more described VT Q_DISTAt least one removed and background ball in vector Humorous coefficient.

Clause 133149-17A.The device of any one of clause 133149-15A or clause 133149-16A, wherein institute It states one or more processors and is further configured to and intermediate different spherical harmonic coefficient is added to after background spherical harmonic coefficient to back Scape spherical harmonic coefficient carries out audio coding, and generates bit stream with the background spherical harmonic coefficient comprising audio coding.

Clause 133149-18A.The device of clause 133149-1A, wherein one or more described processors are further configured To perform the following operation: executing singular value decomposition relative to the multiple spherical harmonic coefficient indicates the multiple spherical harmonic coefficient to generate Left singular vector U matrix, indicate the multiple spherical harmonic coefficient singular value s-matrix and indicate the multiple spherical harmonic coefficient Right singular vector V matrix;Determine one or more V of the transposition of V matrixT DISTVector and one or more VT BGVector, described one Or multiple VDISTOne or more distinct components of vector description sound field, and one or more described VBGVector description sound field it is one or more A background component;Quantify one or more described VT DISTVector is to generate one or more VT Q_DISTVector;And to it is described one or more VT Q_DISTVector reorders to generate one or more V of rearranged sequenceT Q_DISTVector, and wherein one or more described vectors include One or more V of rearranged sequenceT Q_DISTVector.

Figure 40 F is that the various aspects of technology described in the present invention can be performed with packed description two dimension or three dimensional sound in explanation The block diagram of the example audio code device 510F of the spherical harmonic coefficient of field.Audio coding apparatus 510F can be similar to audio coding apparatus 510C, wherein audio coding apparatus 510F includes audio compressing unit 512, audio coding unit 514 and bitstream producing unit 516.In addition, the audio compressing unit 512 of audio coding apparatus 510F can be similar to the audio compression of audio coding apparatus 510C Unit, wherein audio compressing unit 512 includes that decomposition unit 518 and vector reorder unit 532, and the unit can be similar to sound The same unit of frequency code device 510C operates.In some instances, audio coding apparatus 510F may include quantifying unit 534 (as relative to described by Figure 40 D and 40E) is to quantify UDISTVector 525C, UBGVector 525D, VT DISTVector 525E and VT BGVector One or more vectors of any one of 525J.

However, the audio compressing unit 512 of audio coding apparatus 510F may differ from the audio of audio coding apparatus 510C Compression unit 512, the difference is that 524 executable content of protrusion Component Analysis unit of sound field component extraction unit 520 point Analyse the number (being expressed as D in the context of Figure 40 A to 40J) to select prospect component.In other words, prominent Component Analysis list Member 524 can be operated in a manner of as described above relative to U, S and V matrix 519 is with the version of being decomposed for identifying spherical harmonic coefficient It is still generated from Composite tone object from using the natural record of microphone.Prominent Component Analysis unit 524 can then be based on this Synthesis determines and determines D.

In addition, the audio compressing unit 512 of audio coding apparatus 510F may differ from the audio of audio coding apparatus 510C Compression unit 512, the difference is that sound field component extraction unit 520 may include that additional unit, i.e. Degree reduction and energy are protected Stay unit 528F (explanation is " Degree reduction and energy stick unit 528F ").For this reason, audio coding apparatus 510F Sound field component extraction unit 520 is expressed as " sound field component extraction unit 520F ".

Degree reduction and energy stick unit 528F indicate to be configured to the right surprise to the multiple spherical harmonic coefficient 511 is indicated The V of incorgruous amountBGThe background component of matrix 525H executes Degree reduction and retains simultaneously partly by complete VBGMatrix 525H description The unit of the total energy (and adjoint acoustic pressure) of sound field.In in this respect, Degree reduction and energy stick unit 528F can be held 38 institute of Foreground selection unit 48 and energy compensating unit of audio coding apparatus 20 shown in example of the row above in relation to Fig. 4 The operation of description.

Complete VBGMatrix 525H has dimension (N+1)2×(N+1)2- D, wherein D expression are determined as the different of sound field The number of audio component aspect principal component outstanding or (in other words) singular value.That is, complete VBGMatrix 525H includes through determination For those of the background (BG) or (in other words) environment of sound field or non-different audio component singular value.

As described by the Degree reduction unit 524 above in relation to such as Figure 40 B to 40E, Degree reduction and energy retain Unit 528F can be removed, eliminate or delete in other ways (often through pulverised) VBGPair in the background singular value of matrix 525H It should be in those of higher-order spherical shape basis function background singular value.Degree reduction and energy stick unit 528F can be by VBGMatrix The version of being reduced of 525H (is expressed as " VBG'Matrix 525I ", and hereinafter referred to as " through reducing VBG'Matrix 525I ") it is output to Transposition unit 522.Through reducing VBG'Matrix 525I can have dimension (η+1)2×(N+1)2- D, wherein η < N.Transposition unit 522 will Transposition operation is applied to through reducing VBG'Matrix 525I is to generate through transposition through reducing VT BG'Matrix 525J and it is output to mathematics Unit 526, mathematical units 526 can be operated by using UBGMatrix 525D, SBGMatrix 525B and through transposition through reduce VT BG' Matrix 525J calculates UBG*SBG*VT BGAnd rebuild the background sound component of sound field.

According to the techniques described herein, Degree reduction and energy stick unit 528F are further configured to compensate for sound field The total energy of background sound component may be decreased, and described may be decreased is by reducing complete VBGThe order of matrix 525H is to produce Life is through reducing VBG'Matrix 525I and cause.In some instances, Degree reduction and energy stick unit 528F, which pass through, determines to be in put The compensating gain of big value form and compensate, the compensating gain will be applied to through reduction VBG'(N+1) of matrix 525I2-D Column each of so as to will through reduce VBG'Root mean square (RMS) energy of matrix 525I increase to be equal to or at least closer in Complete VBGThe RMS of matrix 525H, then will be through reducing VBG'Matrix 525I is output to transposition unit 522.

In some cases, Degree reduction and energy stick unit 528F can determine complete VBGEach column of matrix 525H RMS energy and through reduce VBG'The RMS energy of each column of matrix 525I, then by the value of magnification of the column be determined as the former with The ratio of the latter, as indicated by following equation:

∝=vBG/vBG',

Wherein ∝ is the value of magnification for a column, vBGIndicate VBGThe single column of matrix 525H, and vBG'Indicate VBG'Matrix The correspondence single row of 525I.This can be indicated by matrix notation are as follows:

A=VBG RMS/VBG'RMS,

Wherein VBG RMSIt is to have to indicate VBGThe RMS vector of the element of the RMS of each column of matrix 525H, VBG'RMSIt is to have It indicates through reducing VBG'The RMS vector of the element of the RMS of each column of matrix 525I, and A is that have for VBGMatrix 525H's is every The value of magnification vector of the element of one column.Degree reduction and energy stick unit 528F are using corresponding value of magnification ∝ or in the form of vectors Scalar multiplication is applied to through reducing VBGEach column of matrix 525I:

V″BG=VBG'AT,

Wherein V "BGIt indicates to include energy compensating through reducing VBG'Matrix 525I.Degree reduction and energy stick unit 528F Can by comprising energy compensating through reducing VBG'Matrix 525I is output to transposition unit 522 so that through reducing VBG'Matrix 525I's RMS and complete VBGThe RMS of matrix 525H balanced (or almost balanced).Comprising energy compensating through reduce VBG'Matrix 525I's is defeated Dimension can be (η+1) out2×(N+1)2-D。

In some instances, to be determining through reducing VBG'Matrix 525I and complete VBGThe respective column of matrix 525H it is each RMS, Degree reduction and energy stick unit 528F will can be applied to the column with reference to spherical harmonic coefficient (SHC) reconstructor first.It is logical It crosses Degree reduction and energy stick unit 528F is applied and determined RMS to determine by through contracting in the domain SHC with reference to the permission of SHC reconstructor Subtract VBG'Matrix 525I and complete VBGThe energy of the overall sound field of each column description of frame represented by matrix 525H.Thus, herein In a little examples, Degree reduction and energy stick unit 528F can will be applied to complete V with reference to SHC reconstructorBGMatrix 525H's is every One arranges and is applied to through reducing VBG'The each of matrix 525I arranges through reduction, determines the column and the corresponding RMS value through reduction column, And the value of magnification for being used for the column is determined as to the RMS value of the column and the ratio of the RMS value through reduction column.In some realities In example, reach through reducing VBG'The Degree reduction of matrix 525I consistently carries out by column with energy reservation.This can be expressed in following puppet In code:

R=refers to reconstructor;

For m=numDist+1:numChanncls

FullV=V (:, m);// take one of V to arrange=> fullV

ReducedV=[fullV (l:numBG);zeros(numChannels-numBG,l)];

Alpha=sqrt (sum ((fullV'*R) ^2)/sum ((reducedV'*R) .^2));

If isnan (alpha) | | isinf (alpha), alpha=1;end;

V_out (:, m)=reducedV*alpha;

end

In the above pseudo-code, numChannels can indicate (N+1)2- D, numBG can indicate (η+1)2, V can indicate VBGSquare Battle array 525H, and V_out can be indicated through reducing VBG'Matrix 525I, and R can indicate the ginseng of Degree reduction and energy stick unit 528F Examine SHC reconstructor.The dimension of V can be (N+1)2×(N+1)2- D, and the dimension of V_out can be (η+1)2×(N+1)2-D。

As a result, audio coding apparatus 510F can be used comprising since order contracts when indicating the multiple spherical harmonic coefficient 511 Subtract process and can the loss of energy energy compensation the reduced V of orderBG'Matrix 525I rebuilds background sound component.

Figure 40 G is that the various aspects of technology described in the present invention can be performed with packed description two dimension or three dimensional sound in explanation The block diagram of the example audio code device 510G of the spherical harmonic coefficient of field.In the example of Figure 40 G, audio coding apparatus 510G includes Sound field component extraction unit 520F.Also, sound field component extraction unit 520F includes prominent Component Analysis unit 524G.

However, the audio compressing unit 512 of audio coding apparatus 510G may differ from the audio pressure of audio coding apparatus 10F Contracting unit 512, the difference is that the audio compressing unit 512 of audio coding apparatus 510G includes prominent Component Analysis unit 524G.Prominent Component Analysis unit 524G can indicate to be configured to using the information based on directionality associated with audio data Indicate the high-lighting of the audio data of sound field or the unit of diversity.

Although the determination based on energy can improve and identify the different audio point of sound field by the reproduction of the SVD sound field decomposed Amount, but in the case where background audio component shows high level, it can also make device mistakenly by background sound based on the determination of energy Frequency component is identified as different audio component.That is, the different and background audio component separation based on only energy can be unstable, because High energy (for example, loudspeaker) background audio component may be improperly identified as different audio component.In order to more steadily exist The different of sound field is distinguished between background audio component, and the various aspects of technology described in the present invention can make prominent component point Analysis unit 524G be able to carry out SHC 511 based on the analysis of directionality with by different and background audio component and SHC 511 Decomposed version separation.

In the example of Figure 40 H, prominent Component Analysis unit 524G can be indicated to be configured or in addition can operate with will be different (or prospect) element with include that background element in one or more of V matrix 519, s-matrix 519B and U matrix 519C separates Unit, similar to previously described audio coding apparatus 510 arrive 510F protrusion Component Analysis unit 524.According to some Technology based on SVD, can component (for example, V, S and U matrix 519 in 519C or the matrix being derived from one or Preceding a small number of vectors of more persons) it handles as distinct components.However, matrix 519 to one or more of 519C can component (its Indicated by vector) most directive component/signal can not be indicated in all situations.

Different from previously described prominent Component Analysis unit 524, prominent implementable this paper of Component Analysis unit 524G Described in technology one or more aspects based on one or more of matrix 519 to 519C or the matrix being derived from The directionality of vector identifies foreground elements.In some instances, prominent Component Analysis unit 524G can be based on energy and vector Directionality identifies or selects one or more vectors as different audio component (wherein component is also known as " object ").Citing comes It says, prominent Component Analysis unit 524G can be by the display of one or more of matrix 519 to 519C (or the matrix being derived from) Those of high-energy and high directivity (for example, being expressed as directionality quotient) vector are identified as different audio component.As a result, if prominent Out Component Analysis unit 524G determine when with one or more of matrix 519 to 519C (or the matrix being derived from) it is other to One specific vector is relatively small when amount compares, then regardless of energy level associated with the specific vector, prominent Component Analysis unit 524G can determine that the specific vector indicates background (or environment) audio component by the sound field indicated of SHC 511.In this respect In, audio coding apparatus 20 shown in the prominent executable example being similar to above in regard to Fig. 4 of Component Analysis unit 524G Analysis of The Acoustic Fields unit 44 describe operation operation.

In some embodiments, prominent Component Analysis unit 524G can be identified by executing following operation based on directionality Different audio object (as noted before, to be also known as " component ").Prominent Component Analysis unit 524G can be (for example, use One or more matrix multiplication processes) by V matrix 519A multiplied by s-matrix 519B.By the way that V matrix 519A is multiplied with s-matrix 519B, Prominent Component Analysis unit 524G can get VS matrix.In addition, prominent Component Analysis unit 524G can be by vector (its of VS matrix Can be row) each of at least some of entry involution (that is, pressing quadratic power exponentiation).In some cases, prominent point Those square of entry associated with the order greater than 1 of each vector can be summed by measuring analytical unit 524G.As a reality Example, if each vector of matrix includes 25 entries, prominent Component Analysis unit 524G can will be opened relative to each vector It starts from fifth entry and ends at the entry involution of each vector of Article 25 purpose, square entry is summed to determine direction Property quotient (or directional indicator).Each summation operation can lead to the directionality quotient of corresponding vector.In this example, prominent point Amount analytical unit 524G can determine those entries associated with the order less than or equal to 1 of every a line (that is, first to the 4th Entry) more generally it is directed to the amount of energy, and the less directionality for those entries.That is, associated with the order of zero or one Lower-order ambiophony correspond to spherical substrate function, such as illustrated in fig. 1 and fig. 2, the ball for the direction of pressure wave Face basis function does not provide many, and is to provide some volume (it indicates energy).

Operating described in above example can also express according to following pseudo-code.Following pseudo-code include mark, in comprising The form of comment statement in character string "/* " and the continuous example of " */" (no quotation marks).

[U, S, V]=svd (audioframe, ' ecom');

VS=V*S;

/ * next line be for independently analyzing every a line, and from fifth entry to Article 25 mesh to first (as One example) value summation in row to be to determine directionality quotient or the directivity measure of corresponding vector.Before summing to the item Mesh is squared.In every a line with greater than 1 the associated entry of order it is associated with higher-order ambiophony, and thus more may be used There can be directionality.*/

SumVS=sum (VS (5:end :) .^2,1);

/ * next line be for the VS matrix that will generate square summation classification, and selection maximum value set (for example, Three in maximum value or four) */

[~, idxVS]=sort (sumVS, ' descend');

U=U (:, idxVS);

V=V (:, idxVS);

S=S (idxVS, idxVS);

In other words, according to the above pseudo-code, prominent Component Analysis unit 524G be may be selected from corresponding to greater than one The entry of each vector for the VS matrix that those of SHC 511 of spherical substrate function of order is decomposed.Prominent Component Analysis list First 524G can will be identified, be calculated through square entry summation or separately then by these entry involutions of each vector of VS matrix The directionality of each vector of outer determining VS matrix is measured or quotient.Next, prominent Component Analysis unit 524G can be based on vector Each of the measurement of respective party tropism the vector of VS matrix is classified.Prominent Component Analysis unit 524G can be by directionality amount The descending of degree classifies these vectors, so that having those of the corresponding directionality of highest vector is first, and with minimum right Those of directionality answered vector is last.What prominent Component Analysis unit 524G can then select vector has highest opposite side The non-zero subset of tropism measurement.

According to some aspects of technology described herein, audio coding apparatus 510G or its one or more component can be incited somebody to action The predetermined number vector of VS matrix is identified as or additionally may act as different audio component.For example, VS matrix is being selected The entry 5 to 25 of every a line and selected entry is summed with after the measurement of the opposite side tropism of each corresponding vector of determination, it is prominent Further selection between the implementable vector of Component Analysis unit 524G is to identify the vector for indicating different audio component.One In a little examples, prominent Component Analysis unit 524G can select the predetermined number of VS matrix by comparing the directionality quotient of vector Vector.As an example, it (and is categorized that tool, which may be selected, there are four highest directionality quotient in prominent Component Analysis unit 524G Four vectors first of VS matrix) four indicated in VS matrix vector.Also, prominent Component Analysis unit 524G can determine Four selected vectors indicate to indicate associated four most different audio objects with the corresponding SHC of sound field.

In some instances, the prominent rearrangeable sequence of Component Analysis unit 524G vector derived from VS matrix, to reflect four The diversity of a selected vector, as described above.In an example, described in the prominent rearrangeable sequence of Component Analysis unit 524G to Amount, so that four selected entries are through being repositioned onto the top of VS matrix.For example, prominent Component Analysis unit 524G can be repaired Change VS matrix, so that all four selected entries are positioned in first (or highest) row of the VS matrix of the rearranged sequence of gained.Though So herein in relation to prominent Component Analysis unit 524G description, but in various embodiments, audio coding apparatus 510G's It reorders described in other components (unit 532 for example, vector reorders) are executable.

Prominent Component Analysis unit 524G can transmit gained matrix (that is, rearranged sequence or VS matrix without reordering) To bitstream producing unit 516.Also, VS matrix 525K can be used to generate bit stream 517 in bitstream producing unit 516.For example, such as Prominent Component Analysis unit 524G's VS matrix 525K of fruit has reordered, then bitstream producing unit 516 can be by the warp of VS matrix 525K The top row for the version that reorders is used as different audio object, such as the rearranged sequence version by quantifying or abandoning VS matrix 525K Residual vector.By quantifying the residual vector of the rearranged sequence version of VS matrix 525K, bitstream producing unit 16 can will be remaining Vector is as environment or background audio data processing.

It is wherein protruding in the example that Component Analysis unit 524G not yet reorders VS matrix 525K, bitstream producing unit 516 can be based on such as by the particular items of every a line of the VS matrix 525K of prominent Component Analysis unit 524G selection (for example, the 5th To 25 entries) and distinguish different audio data and background audio data.For example, bitstream producing unit 516 can pass through quantization Or it abandons the first four entry of every a line of VS matrix 525K and generates bit stream 517.

By this method, audio coding apparatus 510G and/or its component (for example, prominent Component Analysis unit 524G) are implementable Technology of the invention is with the ratio of the energy of higher and lower coefficient determining or furthermore with audio data, so as in Xiang Yiyin It is distinguished between frequency object and the background audio data for indicating sound field.For example, as described, prominent Component Analysis unit 524G Using the energy ratio of the value of the various entries based on the VS matrix 525K generated by prominent Component Analysis unit 524H.Pass through group The data provided by V matrix 519A and s-matrix 519B are closed, prominent Component Analysis unit 524G can produce VS matrix 525K to provide The directionality of various components about audio data and the information of gross energy, in vector and related data (for example, directionality Quotient) form.More specifically, V matrix 519A, which can provide, determines related information with directionality, and s-matrix 519B can provide Related information is determined with the gross energy of the component for audio data.

In other examples, the V to reorder can be used in prominent Component Analysis unit 524GT DISTVector 539 generates VS matrix 525K.In these examples, prominent Component Analysis unit 524G can be based on V matrix before any modification based on s-matrix 519B 519 determine diversity.In other words, according to these examples, the prominent usable only V matrix 519 of Component Analysis unit 524G is determined Directionality, without executing the step of generating VS matrix 525K.More specifically, V matrix 519A can provide about mixed audio number According to component (for example, vector of V matrix 519) mode information, and potentially, about the various of the data conveyed by vector The information of synergy.For example, V matrix 519A can provide such as byIt is relayed to audio coding apparatus 510G " arrival direction " about the various audio components indicated by vector information, such as the arrival direction of each audio component. As used herein, term " component of audio data " can be with any one of matrix 519 or any matrix being derived from " entry " be used interchangeably.

Some embodiments of technology according to the present invention, prominent Component Analysis unit 524G can indicate to mend to SHC It fills or enhances external information to carry out various determinations described herein.As an example, prominent Component Analysis unit 524G External information can be enhanced SHC to determine the high-lighting of the various audio components indicated in matrix 519 into 519C.As another One example, prominent Component Analysis unit 524G and/or vector reorder unit 532 can be to HOA enhancing external data different It is distinguished between audio object and background audio data.

In some instances, prominent Component Analysis unit 524G can detect the part of audio data (for example, different audio pair As) display keynes (Keynesian) energy.The example of these different objects can be associated with the Human voice of modulation.It is adjusting In the case where the audio data based on speech of system, prominent Component Analysis unit 524G be can determine as the energy with residual components Ratio modulation data energy pass by any time keep it is constant (for example, constant in a threshold range) or substantially solid It is fixed.Traditionally, if the energy response of the different audio component with keynes energy is (for example, associated with modulation speech Those energy responses) change between audio frame, then sequence of audio component may not be identified as individual signals by device. However, the prominent implementable technology of the invention of Component Analysis unit 524G is with the distance of the determining vector being expressed as in various matrixes The directionality of object or aperture.

More specifically, prominent Component Analysis unit 524G can determine that the characteristic such as directionality and/or aperture less may be used Can substantially it change across audio frame.As used herein, aperture indicates the coefficient of higher order in audio data to lower level Several ratios.Every a line of V matrix 519A may include the vector corresponding to specific SHC.Prominent Component Analysis unit 524G can be true Determine lower-order SHC (for example, associated with the order less than or equal to 1) to tend to indicate environmental data, and higher-order entry is inclined To in expression differing data.In addition, prominent Component Analysis unit 524G can determine in many cases, higher-order SHC (for example, It is associated with the order greater than 1) display large energy, and in audio interframe, the energy ratio of higher-order SHC and lower-order SHC It keeps substantially like (or constant).

The characteristic of the determining audio data of V matrix 519, example can be used in one or more components of prominent Component Analysis unit 524G Such as directionality and aperture.By this method, the component (for example, prominent Component Analysis unit 524G) of audio coding apparatus 510G can be real Technology described herein is applied to use the information based on directionality to determine high-lighting and/or by different audio object and background Audio distinguishes.High-lighting and/or diversity are determined by using directionality, prominent Component Analysis unit 524G is comparable to be configured To use the data for being based only upon energy to determine, the device of high-lighting and/or diversity realizes more steady determination.Although the above phase The description that determines based on directionality for directionality and/or diversity, but implementable hair of prominent Component Analysis unit 524G Bright technology determines the prominent of the certain components of audio data to go back user's tropism other than other characteristics (for example, energy) Property and/or diversity out, such as by the vector table of one or more of matrix 519 to 519C (or any matrix being derived from) Show.

In some instances, a kind of method include based on the directionality determined for one or more of audio object from One or more associated spherical harmonic coefficients (SHC) of the audio object identify one or more different audio objects.In an example In, the method, which is further included, determines one or more audio objects based on spherical harmonic coefficient associated with audio object Directionality.In some instances, the method, which further includes, executes the singular value decomposition relative to spherical harmonic coefficient to generate table Show the s-matrix and table of the U matrix of the left singular vector of the multiple spherical harmonic coefficient, the singular value for indicating the multiple spherical harmonic coefficient Show the V matrix of the right singular vector of the multiple spherical harmonic coefficient;And by the multiple spherical harmonic coefficient be expressed as U matrix, s-matrix and At least part of function of one or more of V matrix, wherein determining the respective party tropism of one or more audio objects It is to be at least partially based on V matrix.

In an example, the method, which further includes, reorders one or more vectors of V matrix so that through weight In the V matrix of sequence, the vector with larger directionality quotient is positioned above the vector with smaller directionality quotient.In a reality In example, the method, which further includes, determines that the vector with larger directionality quotient includes than the vector with smaller directionality quotient Big directional information.In an example, the method is further included V Matrix Multiplication with s-matrix to generate VS matrix, described VS matrix includes one or more vectors.In an example, the method further include selection VS matrix every a line with Greater than the associated entry of 1 order, by each of selected entry certainly multiplied by corresponding square of entry of formation, and it is directed to VS All squares of entry summations are determined the directionality quotient of corresponding vector by every a line of matrix.

In some instances, every a line of VS matrix includes 25 entries.In an example, each of VS matrix is selected Capable entry associated with the order greater than 1 includes the 5th entry of every a line that selection starts from VS matrix and ends at VS All entries of 25th entry of every a line of matrix.In an example, the method further includes selection VS matrix The subset of vector indicates different audio object.In some instances, select the subset include selection four of VS matrix to Amount, and selected four vectors have four maximum directivity quotient of institute's directed quantity of VS matrix.In an example, vector is determined Selection subsets indicate that different audio object is both directionality and energy based on each vector.

In some instances, a kind of method includes based on the directionality and energy determined for one or more of audio object Amount identifies one or more different audio objects from one or more spherical harmonic coefficients associated with the audio object.In an example In, the method, which is further included, determines one or more audio objects based on spherical harmonic coefficient associated with audio object One or both in directionality and energy.In some instances, the method, which further includes, executes relative to expression sound field The singular value decomposition of spherical harmonic coefficient is to generate the U matrix for the left singular vector for indicating the multiple spherical harmonic coefficient, indicate described more The V matrix of the right singular vector of the s-matrix and the multiple spherical harmonic coefficient of expression of the singular value of a spherical harmonic coefficient;And it will be described more A spherical harmonic coefficient is expressed as at least part of function of one or more of U matrix, s-matrix and V matrix, wherein described in determining The respective party tropism of one or more audio objects is to be at least partially based on V matrix, and wherein determine one or more described audios pair The corresponding energy of elephant is to be at least partially based on s-matrix.

In an example, the method is further included V Matrix Multiplication with s-matrix to generate VS matrix, the VS square Battle array includes one or more vectors.In some instances, the method further include selection VS matrix every a line be greater than 1 The associated entry of order, by each of selected entry from multiplied by formed corresponding square of entry, and be directed to VS matrix Every a line, the summation of all squares of entries is generated to the directionality quotient of the correspondence vector of VS matrix.In some instances, VS square Every a line of battle array includes 25 entries.In an example, the associated with the order greater than 1 of every a line of VS matrix is selected Entry includes the institute of the 25th entry of the every a line for selecting the 5th entry of the every a line for starting from VS matrix and ending at VS matrix There is entry.In some instances, the method further includes the subset of selection vector to indicate different audio object.At one In example, selecting the subset includes four vectors for selecting VS matrix, and selected four vectors have the institute of VS matrix oriented Four maximum directivity quotient of amount.In some instances, determine that the selection subsets of vector indicate that different audio object is based on every Both the directionality of one vector and energy.

In some instances, a kind of method includes that different point that describes the sound field is determined using the information based on directionality One or more secondary vectors of the background component of one or more primary vectors and description sound field of amount, it is described one or more the Both one vector and one or more secondary vectors be at least through execute relative to the multiple spherical harmonic coefficient transformation and It generates.In an example, the transformation includes singular value decomposition, and singular value decomposition, which generates, indicates the multiple spherical harmonic coefficient The U matrix of left singular vector, indicate the multiple spherical harmonic coefficient singular value s-matrix and indicate the multiple spherical harmonic coefficient The V matrix of right singular vector.In an example, transformation includes principal component analysis to identify the distinct components and sound field of sound field Background component.

In some instances, a kind of device is configured or in addition can operate to execute any in techniques described herein Any combination of person or the technology.In some instances, a kind of computer-readable storage medium is encoded instruction, the finger Enable any group for causing one or more processors to execute any one of techniques described herein or the technology when implemented It closes.In some instances, a kind of device includes to execute any of the technologies described herein or the technology The device of any combination.

That is, the aforementioned aspects of the technology can enable audio coding apparatus 510G be configured to be grasped according to following clause Make.

Clause 134954-1B.A kind of device, such as audio coding apparatus 510G comprising: one or more processors, It is configured to based on the directionality and Thin interbed determined for one or more of audio object from related to audio object One or more different audio objects of one or more spherical harmonic coefficients of connection.

Clause 134954-2B.The device of clause 134954-1B, wherein one or more described processors are further configured One in directionality and energy to determine one or more audio objects based on spherical harmonic coefficient associated with audio object Or both.

Clause 134954-3B.Such as the device of any one of technical solution 1B or 2B or combinations thereof, wherein it is described one or more Processor, which is further configured to execute, indicates described more relative to the singular value decomposition for the spherical harmonic coefficient for indicating sound field to generate The U matrix of the left singular vector of a spherical harmonic coefficient, the s-matrix of the singular value of the multiple spherical harmonic coefficient of expression and expression are described more The V matrix of the right singular vector of a spherical harmonic coefficient, and the multiple spherical harmonic coefficient is expressed as in U matrix, s-matrix and V matrix One or more at least part of function, wherein one or more described processors are configured to be at least partially based on V matrix Determine the respective party tropism of one or more audio objects, and wherein one or more described processors be configured to determine it is described The corresponding energy of one or more audio objects is to be at least partially based on s-matrix.

Clause 134954-4B.The device of clause 134954-3B, wherein one or more described processors are further configured With by V Matrix Multiplication with s-matrix to generate VS matrix, the VS matrix includes one or more vectors.

Clause 134954-5B.The device of clause 134954-4B, wherein one or more described processors are further configured To select the entry associated with the order greater than 1 of every a line of VS matrix, by each of selected entry from multiplied by shape At corresponding square of entry, and be directed to every a line of VS matrix, by the summation of all squares of entries with generate the correspondence of VS matrix to The directionality quotient of amount.

Clause 134954-6B.Such as the device of any one of technical solution 5B and 6B or combinations thereof, wherein VS matrix is each Row includes 25 entries.

Clause 134954-7B.The device of clause 134954-6B, wherein one or more described processors are configured to select Start from the 5th entry of every a line of VS matrix and ends at all entries of the 25th entry of every a line of VS matrix.

Clause 134954-8B.The device of any one of clause 134954-6B and clause 134954-7B or combinations thereof, wherein One or more described processors are further configured to select the subset of vector to indicate different audio object.

Clause 134954-9B.The device of clause 134954-8B, wherein one or more described processors are configured to select Four vectors of VS matrix, and wherein select four maximum directivity quotient that four vectors have institute's directed quantity of VS matrix.

Clause 134954-10B.The device of any one of clause 134954-8B and clause 134954-9B or combinations thereof, Described in one or more processors be further configured to determine that the selection subsets of vector indicate that different audio object is based on every Both the directionality of one vector and energy.

Clause 134954-1C.A kind of device, such as audio coding apparatus 510G comprising: one or more processors, institute One or more processors are stated to be configured to determine one or more for describing the distinct components of sound field using the information based on directionality Primary vector and describe sound field background component one or more secondary vectors, one or more described primary vectors and described one or Both multiple secondary vectors are generated at least through the transformation relative to the multiple spherical harmonic coefficient is executed.

Clause 134954-2C.The method of clause 134954-1C, wherein the transformation includes singular value decomposition, it is described unusual Value, which is decomposed, to be generated the U matrix for the left singular vector for indicating the multiple spherical harmonic coefficient, indicates the unusual of the multiple spherical harmonic coefficient The V matrix of the right singular vector of the s-matrix and the multiple spherical harmonic coefficient of expression of value.

Clause 134954-3C.The method of clause 134954-2C further comprises by clause 134954-1A to clause The operation of any combination of 134954-12A and clause 134954-1B to 134954-9B narration.

Clause 134954-4C.The method of clause 134954-1C, wherein the transformation includes principal component analysis to identify sound The distinct components of field and the background component of sound field.

Figure 40 H is that the various aspects of technology described in the present invention can be performed with packed description two dimension or three dimensional sound in explanation The block diagram of the example audio code device 510H of the spherical harmonic coefficient of field.Audio coding apparatus 510H can be similar to audio coding apparatus 510G, because audio coding apparatus 510H includes audio compressing unit 512, audio coding unit 514 and bitstream producing unit 516.In addition, the audio compressing unit 512 of audio coding apparatus 510H can be similar to the audio compression of audio coding apparatus 510G Unit, because audio compressing unit 512 includes decomposition unit 518 and sound field component extraction unit 520G, the unit can be similar It is operated in the same unit of audio coding apparatus 510G.In some instances, audio coding apparatus 510H may include quantifying unit 534 (as relative to described by Figure 40 D to 40E) are to quantify UDISTVector 525C, UBGVector 525D, VT DISTVector 525E and VT BG One or more vectors of any one of vector 525J.

However, the audio compressing unit 512 of audio coding apparatus 510H may differ from the audio of audio coding apparatus 510G Compression unit 512, the difference is that, the audio compressing unit 512 of audio coding apparatus 510H includes to be expressed as interpolation list The additional unit of member 550.Interpolation unit 550 can indicate the first audio frame of subframe of the interpolation from the first audio frame subframe and The unit of second time upper subsequent or previous audio frame, is such as more fully described below in relation to Figure 45 and 45B.Interpolation unit 550 can reduce meter by potentially reducing the degree for needing decomposition unit 518 to decompose SHC 511 in executing this interpolation process It calculates complexity (for processing cycle and/or memory consumption).In in this respect, interpolation unit 550 is executable be similar to On relative to the audio coding apparatus 24 shown in the example in figure 4 space-time interpolation unit 50 describe operation behaviour Make.

That is, the singular value decomposition executed by decomposition unit 518 potentially there is processor and/or memory to concentrate very much Property, while again in some instances, increase especially as the order of SHC 511, takes a significant amount of time to decompose SHC 511.For It reduces time quantum and makes the compression of SHC 511 more efficiently (for processing cycle and/or memory consumption), in the present invention Described technology can provide the interpolation of one or more subframes of the first audio frame, and wherein each of subframe can indicate SHC 511 through decompose version.Substitution executes SVD relative to entire frame, and the technology can enable decomposition unit 518 decompose the first sound First subframe of frequency frame, to generate V matrix 519'.

Second subframe of also decomposable second audio frame of decomposition unit 518, wherein this second audio frame can be on the time the After one audio frame or on the time before the first audio frame.Decomposition unit 518 can export V for this subframe of the second audio frame Matrix 519'.Interpolation unit 550 can be then based on V matrix 519' the first audio frame of interpolation decomposed from first and second subframe Remaining subframe exports V matrix 519, s-matrix 519B and U matrix 519C, wherein can the V square based on SHC 511, the first audio frame Battle array 519A and the interpolated V matrix 519 of the remaining subframe for the first audio frame calculate the decomposition for remaining subframe.Interpolation can Therefore the calculating of the decomposition of the remaining subframe for the first audio frame is avoided.

In addition, as noted before, U matrix 519C can not be continuous in interframe, wherein can with from the second of SHC 511 The U matrix decomposed from the first audio frame of SHC 511 is specified in the U matrix 519C that audio frame decomposes in different row and/or column The distinct components of 519C.By executing this interpolation, if linear interpolation, which can have reduce, is attributed to frame boundaries (or in other words Say, SHC 511 arrives the segmentation of frame) and the smoothing effect of any illusion of introducing, then it can reduce discontinuity.Use V matrix 519', which executes this interpolation and then restores U matrix 519C from SHC 511 based on interpolated V matrix 519', can make from the U that reorders Any effect of matrix 519C is smooth.

In operation, interpolation unit 550 can interpolation from one comprising a spherical harmonic coefficient 511 more than first in the first frame Partial first decomposes (for example, V matrix 519') and a part comprising a spherical harmonic coefficient 511 more than second in the second frame Second decomposes one or more subframes of first audio frame of (for example, V matrix 519') to generate and be used for one or more described subframes The interpolated spherical harmonic coefficient through decomposing.

In some instances, first decompose include indicate the right side of the part of more than first a spherical harmonic coefficient 511 it is unusual to First V matrix 519' of amount.Similarly, in some instances, it second decomposes comprising indicating more than second a spherical harmonic coefficient 2nd V matrix 519' of partial right singular vector.

Interpolation unit 550 can be held based on the first V matrix 519' and the 2nd V matrix 519' relative to one or more described subframes Row temporal interpolation.That is, interpolation unit 550 can be based on the V matrix 519' decomposed from the first subframe of the first audio frame and from second The V matrix 519' that first subframe of audio frame is decomposed four whole subframes of the interpolation for example from the first audio frame in time Second, third and the 4th subframe.In some instances, it is inserted as linear time interpolation in this time, wherein when the first audio of interpolation Fourth son of the V matrix 519' ratio decomposed when the second subframe of frame from the first subframe of the first audio frame when the first audio frame of interpolation Weight heavier when frame.When interpolation third subframe, V matrix 519' can be equably weighted.When four subframe of interpolation, from second The V matrix 519' that first subframe of audio frame is decomposed is comparable from the important place V matrix 519' that the first subframe of the first audio frame is decomposed Weighting.

In other words, the proximity of one of subframe of the first audio frame to be interpolated is given, then linear time interpolation V matrix 519' can be weighted.For the second subframe to be interpolated, the given V matrix decomposed from the first subframe of the first audio frame The proximity of 519' and the second subframe to be interpolated can then compare the first subframe from the second audio frame to the V matrix 519' The important place matrix 519' V of decomposition weights.For this reason, when being based on V matrix 519' interpolation third subframe, weight can be equivalent. The first subframe of the 4th subframe to be interpolated closer second audio frame compared with the first subframe close to the first audio frame is given, The weight for being then applied to the V matrix 519' decomposed from the first subframe of the second audio frame, which can be greater than, to be applied to from the first audio frame The weight for the V matrix 519' that first subframe is decomposed.

Although in some instances, executing interpolation, a ball more than described first using only the first subframe of each audio frame The part of humorous coefficient may include two in four subframes of a spherical harmonic coefficient 511 more than described first.In these and other example In, the part of a spherical harmonic coefficient 511 more than described second includes two in four subframes of a spherical harmonic coefficient 511 more than described second It is a.

As noted before, single device (for example, audio coding apparatus 510H) can be also to decompose a ball more than described first humorous Interpolation is executed while the part of coefficient to generate the first decomposition of the part of a spherical harmonic coefficient more than described first.At these And in other examples, decomposition unit 518 can decompose the part of a spherical harmonic coefficient more than described second to generate a ball more than described second The second of the part of humorous coefficient is decomposed.Although describing relative to single device, this hair is can be performed in two or more devices Technology described in bright, wherein one of described two devices execute decomposition, and the other of described device is according to this Technology described in invention executes interpolation.

It in other words, can be the ginseng of 3D pressure field based on the humorous 3D audio of ball for the orthogonal basement function on spherical surface Number indicates.The order N of the expression is higher, and spatial resolution is possibly higher, and the number of usually humorous (SH) coefficient of ball is bigger ((N+1) in total2A coefficient).For many applications, it may be necessary to which the bandwidth reduction of coefficient can be transmitted effectively and store institute State coefficient.This targeted in the present invention technology can provide to be reduced using the dimension based on frame of singular value decomposition (SVD) Journey.Each frame of coefficient can be resolved into three matrix Us, S and V by SVD analysis.In some instances, the technology can will be in U Some durection component disposition as basic sound field in vector.However, when disposing by this method, these vectors (in U) Even if being discontinuous in interframe -- it indicates same different audio component.When passing through component described in the feed-in of converting audio frequency decoder When, these discontinuities can lead to significant illusion.

Technology described in the present invention can solve this discontinuity.That is, the technology can be based on following observed result: V Matrix can the interpreted orthogonal intersection space axis in the humorous domain of ball.U matrix can indicate humorous (HOA) data of ball according to those basis functions Projection, wherein discontinuity is attributable to basis function (V), and the basis function (V) changes each frame and therefore itself is not Continuously.This is different from the similar decomposition of such as Fourier transform, and wherein basis function will be in some instances normal in interframe Number.In these terms, SVD is regarded as matching and pursues algorithm.Technology described in the present invention can enable interpolation unit 550 It is enough to maintain the continuity between basis function (V) in interframe and progress interpolation therebetween.

In some instances, the technology enables interpolation unit 550 that the frame of SH data is divided into four subframes, as above It is described and further described below in relation to Figure 45 and 45B.Interpolation unit 550 can then calculate the SVD of the first subframe.It is similar Ground, we calculate the SVD of the first subframe of the second frame.For each of first frame and the second frame, interpolation unit 550 can lead to Space reflection will be transformed into for the vector in V (using the projection matrix of such as T design matrix) on vector projection to spherical surface by crossing.It is interior Can then the vector in V be interpreted as the shape on spherical surface by inserting unit 550.In order in the first subframe of first frame and next frame The first subframe between three subframes of interpolation V matrix, interpolation unit 550 can then these spatial forms of interpolation, and then pass through Inverse by projection matrix is transformed back to SH vector.By this method, technology of the invention can provide smooth turn between V matrix Become.

By this method, audio coding apparatus 510H can be configured to perform the technology illustrated below in relation to following clause Various aspects.

Clause 135054-1A.A kind of device, such as audio coding apparatus 510H comprising: one or more processors, institute It states one or more processors and is configured to of interpolation from a part comprising a spherical harmonic coefficient more than first in the first frame One decompose and second first frame decomposed of a part comprising a spherical harmonic coefficient more than second in the second frame one or more Subframe is to generate the interpolated spherical harmonic coefficient through decomposing for one or more subframes.

Clause 135054-2A.The device of clause 135054-1A, wherein described first decomposes including indicating described more than first First V matrix of the right singular vector of the part of a spherical harmonic coefficient.

Clause 135054-3A.The device of clause 135054-1A, wherein described second decomposes including indicating described more than second 2nd V matrix of the right singular vector of the part of a spherical harmonic coefficient.

Clause 135054-4A.The device of clause 135054-1A, wherein first decomposes including indicating more than first a ball First V matrix of the right singular vector of the part of humorous coefficient, and wherein second decompose comprising indicating more than second a humorous system of ball 2nd V matrix of the right singular vector of several parts.

Clause 135054-5A.The device of clause 135054-1A, wherein one or more described processors are further configured To be decomposed based on first and second decompose one or more subframes described in interpolation in time when one or more subframes of interpolation.

Clause 135054-6A.The device of clause 135054-1A, wherein one or more described processors are further configured First to be decomposed and is projected to generate the first Projective decomposition in spatial domain, second is decomposed when one or more subframe of interpolation It projects to generate the second Projective decomposition in spatial domain, spatially the first Projective decomposition of interpolation and the second Projective decomposition are to generate The Projective decomposition of one spatial interpolation and the Projective decomposition of second space interpolation, and the Projective decomposition based on the first spatial interpolation and The Projective decomposition of two spatial interpolations one or more subframes described in interpolation in time.

Clause 135054-7A.The device of clause 135054-6A, wherein one or more described processors are further configured The spherical harmonic coefficient of the time generated from one or more subframes described in interpolation upper interpolation is projected back to the humorous domain of ball.

Clause 135054-8A.The device of clause 135054-1A, wherein the part of a spherical harmonic coefficient more than described first Single subframe including more than described first a spherical harmonic coefficients.

Clause 135054-9A.The device of clause 135054-1A, wherein the part of a spherical harmonic coefficient more than described second Single subframe including more than described second a spherical harmonic coefficients.

Clause 135054-10A.The device of clause 135054-1A,

Wherein the first frame is divided into four subframes, and

Wherein the part of a spherical harmonic coefficient more than described first only includes described of a spherical harmonic coefficient more than described first One subframe.

Clause 135054-11A.The device of clause 135054-1A,

Wherein second frame is divided into four subframes, and

Wherein the part of a spherical harmonic coefficient more than described second only includes described of a spherical harmonic coefficient more than described second One subframe.

Clause 135054-12A.The device of clause 135054-1A, wherein the part of a spherical harmonic coefficient more than described first Two in four subframes including more than described first a spherical harmonic coefficients.

Clause 135054-13A.The device of clause 135054-1A, wherein the part of a spherical harmonic coefficient more than described second Two in four subframes including more than described second a spherical harmonic coefficients.

Clause 135054-14A.The device of clause 135054-1A, wherein one or more described processors are further configured The of the part of a spherical harmonic coefficient more than described first is generated to decompose the part of a spherical harmonic coefficient more than described first One decomposes.

Clause 135054-15A.The device of clause 135054-1A, wherein one or more described processors are further configured The of the part of a spherical harmonic coefficient more than described second is generated to decompose the part of a spherical harmonic coefficient more than described second Two decompose.

Clause 135054-16A.The device of clause 135054-1A, wherein one or more described processors are further configured It generates to execute the singular value decomposition relative to the part of more than described first a spherical harmonic coefficients and indicates more than described first Described in the U matrix of the left singular vector of spherical harmonic coefficient, the s-matrix and expression of the singular value of expression more than first a spherical harmonic coefficient The V matrix of the right singular vector of a spherical harmonic coefficient more than first.

Clause 135054-17A.The device of clause 135054-1A, wherein one or more described processors are further configured It generates to execute the singular value decomposition relative to the part of more than described second a spherical harmonic coefficients and indicates more than described second Described in the U matrix of the left singular vector of spherical harmonic coefficient, the s-matrix and expression of the singular value of expression more than second a spherical harmonic coefficient The V matrix of the right singular vector of a spherical harmonic coefficient more than second.

Clause 135054-18A.The device of clause 135054-1A, wherein first and second described multiple spherical harmonic coefficient are respectively Indicate that the plane wave of sound field indicates.

Clause 135054-19A.The device of clause 135054-1A, wherein first and second described multiple spherical harmonic coefficient are respectively Indicate one or more the monophonic audio objects mixed.

Clause 135054-20A.The device of clause 135054-1A, wherein first and second described multiple spherical harmonic coefficient are respectively First and second corresponding spherical harmonic coefficient comprising indicating three-dimensional sound field.

Clause 135054-21A.The device of clause 135054-1A, wherein first and second described multiple spherical harmonic coefficient are respectively Join at least one spherical substrate functional dependence with the order greater than one.

Clause 135054-22A.The device of clause 135054-1A, wherein first and second described multiple spherical harmonic coefficient are respectively Join at least one spherical substrate functional dependence with the order equal to four.

Although above description is to be executed by audio coding apparatus 510H, various audio decoding apparatus 24 and 540 can also be held Row is above in regard to any one of clause 135054-1A to 135054-22A various aspects of technology illustrated.

Figure 40 I is that the various aspects of technology described in the present invention can be performed with packed description two dimension or three dimensional sound in explanation The block diagram of the example audio code device 510I of the spherical harmonic coefficient of field.Audio coding apparatus 510I can be similar to audio coding apparatus 510H, because audio coding apparatus 510I includes audio compressing unit 512, audio coding unit 514 and bitstream producing unit 516.In addition, the audio compressing unit 512 of audio coding apparatus 510I can be similar to the audio compression of audio coding apparatus 510H Unit, because audio compressing unit 512 includes decomposition unit 518 and sound field component extraction unit 520, the unit be can be similar to The same unit of audio coding apparatus 510H operates.In some instances, audio coding apparatus 10I may include quantifying unit 34 (as relative to described by Fig. 3 D to 3E) is to quantify UDIST 25C、UBG 25D、VT DIST25E and VT BGThe one of any one of 25J Or multiple vectors.

Although however, the audio compression of the audio compressing unit 512 and audio coding apparatus 10H of audio coding apparatus 510I Both units 512 include sound field component extraction unit, but the sound field component extraction unit 520I of audio coding apparatus 510I can be wrapped Containing the additional modules for being referred to as V compression unit 552.V compression unit 552 can indicate the spatial component for being configured to compression sound field (that is, being in this example VT DISTOne or more of vector 539) unit.That is, can relative to the SHC singular value decomposition executed SHC (it indicates sound field) is resolved into the energy component indicated by the vector of s-matrix, the time component indicated by U matrix and by V The spatial component that matrix indicates.V compression unit 552 is executable to be similar to above in regard to operation described in quantifying unit 52 Operation.

For purposes of example, it is assumed that VT DISTVector 539 includes two row vectors (its hint sound respectively with 25 elements The quadravalence HOA of field is indicated).Although describing relative to two row vectors, any number vector may include in VT DISTVector It in 539, is at most (n+1)2A, wherein n indicates the order that the HOA of sound field is indicated.

V compression unit 552 can receive VT DISTVector 539, and compression scheme is executed to generate compressed VT DISTVector table Show 539'.This compression scheme, which usually can be related to any of the element for compressing vector or data, can be envisaged compression scheme, and not The example that should be limited to following more detail.

As an example, V compression unit 552 is executable to include one or more of following compression scheme: by VT DISTTo The floating point representation of each element of amount 539 transforms to VT DISTThe integer representation of each element of vector 539, VT DISTVector 539 The uniform quantization of integer representation and VT DISTThe classification and decoding of the quantified integer representation of vector 539.The one of this compression scheme Or it is multiple during each can be controlled by dynamic state of parameters to realize or almost realize (as an example) for gained bit stream 517 targeted bit rates.

It is assumed that VT DISTEach of vector 539 is mutually orthogonal, then VT DISTEach of vector 539 can be translated independently Code.In some instances, as follows described in more detail, it same decoding mode (being defined by various subpatterns) can be used to decode Every VT DISTEach element of vector 539.

Under any circumstance, as noted above, this decoding scheme can be first related to VT DISTEach of vector 539 The floating point representation (it is 32 floating numbers in some instances) of each element be transformed to 16 integer representations.V compression unit 552 can be by by VT DISTEach element of given one in vector 539 is multiplied by 215(in some instances, by moving to right 15 And execute) execute this floating-point to integer transform.

V compression unit 552 can then be executed relative to VT DISTThe all elements of the given one in vector 539 it is equal Even quantization.V compression unit 552 can identify quantization step based on the value for being represented by nbits parameter.V compression unit 552 can base This nbits parameter is dynamically determined in targeted bit rates.V compression unit 552 can determine the quantization of the function as this nbits parameter Step-length.As an example, V compression unit 552 can determine quantization step (being expressed as " residual quantity " or " Δ " in the present invention) For equal to 216-nbits.In this example, if nbits is equal to six, residual quantity is equal to 210, and there are 26A quantification levels.At this In aspect, for vector element v, quantified vector element vqEqual to [v/ Δ], and -2nbits-1<vq<2nbits-1

V compression unit 552 can then execute the classification and remaining decoding of quantified vector element.As an example, V pressure Following equation can be used to give quantified vector element v for one for contracting unit 552qIdentify that classification corresponding to this element is (logical Cross and determine class identifier cid):

V compression unit 552 then can carry out Hoffman decodeng to this classification index cid, while also identification instruction vqIt is positive The sign bits of value or negative value.Next V compression unit 552 can recognize the remnants in this classification.As an example, V pressure Contracting unit 552 can determine this remnants according to following equation:

Remaining=| vq|-2cid-1

V compression unit 552 then can carry out block decoding to this remnants with cid-1 position.

Following instance illustrates the simplified example process of this classification and remaining decoding process.First, it is assumed that nbits be equal to six with So that vq∈[-31,31].Next, assuming following:

cid vq Huffman code for cid 0 0 ‘1’ 1 -1,1 ‘01’ 2 -3,-2,2,3 ‘000’ 3 -7,-6,-5,-4,4,5,6,7 ‘0010’ 4 -15,-14,...,-8,8,...,14,15 ‘00110’ 5 -31,-30,...,-16,16,...,30,31 ‘00111’

Also, assuming following:

cid For remaining block code 0 It is unavailable 1 0,1 2 01,00,10,11 3 011,010,001,000,100,101,110,111 4 0111,0110...,0000,1000,...,1110,1111 5 01111,...,00000,10000,...,11111

Thus, for vq=[6, -17,0,0,3], it may be determined that following:

> > cid=3,5,0,0,2

> > sign=l, 0, x, x, l

> > remaining=2,1, x, x, 1

> > is used for 6 position=' 0010 '+' 1 '+' 10 '

> > is used for -17 position=' 00111 '+' 0 '+' 0001 '

> > is used for 0 position=' 0 '

> > is used for 0 position=' 0 '

> > is used for 3 position=' 000 '+' l '+' l '

The position of > > in total=7+10+1+1+5=24

> > average bit=24/5=4.8

Although being not shown in previous simplified example, V compression unit 552 can be directed to nbits when decoding to cid Different value select different huffman codebooks.In some instances, V compression unit 552 can provide for nbits value 6 ..., 15 different Hoffman decodeng tables.In addition, V compression unit 552 may include for range from 6 ..., 15 different nbits values Each of five different huffman codebooks, 50 huffman codebooks altogether.In in this respect, V compression unit 552 can be wrapped The decoding of the cid in many different statistics situations is adapted to containing multiple and different huffman codebooks.

In order to illustrate V compression unit 552 can be for each of nbits value comprising arriving for decoding vector element one Four the first huffman codebook, the second huffman codebook for decoding vector element five to nine, for decoding vector element nine Or more third huffman codebook.As V to be compressedT DISTOne of vector 539 is not from VT DISTTime in vector 539 Upper subsequent corresponding one is predicted and does not indicate Composite tone object (for example, originally being defined by pulse-code modulation (PCM) audio object One Composite tone object) spatial information when, this first three huffman codebook can be used.V compression unit 552 can in addition for Each of nbits value includes for decoding VT DIST4th huffman codebook of the one in vector 539 (works as VT DIST This one in vector 539 is from VT DISTWhen the time of vector 539 upper subsequent corresponding one is predicted).V compression unit 552 is right In each of nbits value also may include for decoding VT DISTThe one in vector 539 the 5th huffman codebook (when VT DISTWhen this one in vector 539 indicates Composite tone object).It can be for these difference statistics situations (that is, in this example In, nonanticipating and non-synthetic situation, prediction situation and synthesis situation) each of the various huffman codebooks of exploitation.

Following table illustrates huffman table selection and to specify in bit stream so that decompression unit can select appropriate Hough The position of graceful table:

Prediction mode HT information HT table 0 0 HT5 0 1 HT{1,2,3} 1 0 HT4 1 1 HT5

In preceding table, prediction mode (" Pred mode ") indicates whether to perform prediction for current vector, and Huffman Table (" HT information ") indicates additional huffman codebook (or table) information to select one of huffman table one to five.

Following table further illustrates this huffman table selection course (it is assumed that various statistics scenes or situation).

Record Synthesis Without prediction HT{1,2,3} HT5 There is prediction HT4 HT5

In preceding table, " record " column instruction vector indicates decoding situation when recorded audio object, and " synthesis " arranges Indicate decoding situation when vector indicates Composite tone object.The instruction of " no prediction " row does not execute prediction relative to vector element When write decoding situation, and the instruction of " having prediction " row executes decoding situation when predicting relative to vector element.Institute in such table Show, V compression unit 552 selected when vector indicates record audio object and does not execute prediction relative to vector element HT 1, 2,3}.V compression unit 552 is selected when audio object indicates Composite tone object and does not execute and predicts relative to vector element HT5.V compression unit 552 selects HT4 when vector indicates recorded audio object and executes and predicts relative to vector element.V pressure Contracting unit 552 selects HT5 when audio object indicates Composite tone object and executes and predicts relative to vector element.

By this method, the technology can enable audio compression device compress the spatial component of sound field, wherein spatial component It is to be generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

Figure 43 is the figure that V compression unit 552 shown in Figure 40 I is described in more detail.In the example of Figure 43, V compression Unit 552 includes uniform quantization unit 600, nbits unit 602, predicting unit 604,606 (" prediction mode of prediction mode unit Unit 606 "), classification and remaining decoding unit 608 and huffman table selecting unit 610.Uniform quantization unit 600 is indicated through matching Set with execute above in regard to one of spatial component for being expressed as v in the example of Figure 43, (it can indicate VT DISTVector 539 Any one of) described in uniform quantization unit.Nbits unit 602 indicates to be configured to determine nbits parameter or value Unit.

The expression of predicting unit 604 is configured to execute relative to being expressed as v in the example of Figure 43qQuantified space point The unit of the prediction of amount.Predicting unit 604 can be by executing VT DISTElement subtracts V to the current one of vector 539 one by oneT DIST The time of vector 539 goes up subsequent corresponding one to execute prediction.The result of this prediction can be referred to prediction spatial component.

Prediction mode unit 606 can indicate the unit for being configured to selection prediction mode.Huffman table selecting unit 610 can Indicate the unit for being configured to select decoding of the appropriate huffman table for cid.As an example, prediction mode unit 606 And huffman table selecting unit 610 can be operated according to following pseudo-code:

For giving nbits, all huffman tables with nbits are retrieved

B00=0;B01=0;B10=0;B11=0;// initialize to calculate the expection position of every decoding mode

Element number in for m=1:(vector)

// calculate the expection number for being used for the position of vector element v (m)

// without prediction and use huffman table 5

B00=B00+calculate_bits (v (m), HT5);

// without prediction and use huffman table { 1,2,3 }

B01=B01+calculate_bits (v (m), HTq);q in{1,2,3}

// calculate the expection number for being used for the position of prediction residue e (m)

E (m)=v (m)-vp (m);//vp (m): former frame vector element

// have prediction and use huffman table 4

B10=B10+calculate_bits (e (m), HT4);

// have prediction and use huffman table 5

B11=B11+calculate_bits (e (m), HT5);

end

// find out the optimum prediction mode and huffman table for generating minimum position

// optimum prediction mode and huffman table add flag by pflag and Htflag respectively

[Be, id]=min ([B00B01B10B11]);

Switch id

Case 1:pflag=0;HTflag=0;

Case 2:pflag=0;HTflag=1;

Case 3:pflag=1;HTflag=0;

Case 4:pflag=1;HTflag=1;

end

Classification and remaining decoding unit 608 can indicate to be configured to execute in the way of being more fully described above predicted The unit of classification and the remaining decoding of spatial component or quantified spatial component (when prediction deactivates).

As shown in the example of Figure 43, the exportable various parameters of V compression unit 552 or value to contain in bit stream 517 or In auxiliary information (it can itself be the bit stream separated with bit stream 517).It is assumed that specifying the information in bit stream 517, then V compresses Nbits value, prediction mode and huffman table information together with spatial component (can be shown as in the example of Figure 40 I by unit unit Through compression space component 539') be output to bitstream producing unit 516 through compressed version, in this example, the spatial component It can refer to be selected to coding cid, sign bits and the Huffman code through block decoding remnants.All V can be directed toT DISTVector 539 exists Specify nbits value primary in bit stream 517, but can be for the specified prediction mode of each of vector 539 and huffman table information. The part of the bit stream through compressed version of designated space component is showed in the example of Figure 10 B and 10C.

By this method, the various sides of the executable technology illustrated below in relation to following clause of audio coding apparatus 510H Face.

Clause 141541-1A.A kind of device, such as audio coding apparatus 510H comprising: one or more processors, institute One or more processors are stated to be configured to obtain the bit stream through compressed version of the spatial component comprising sound field, the spatial component It is to be generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

Clause 141541-2A.The device of clause 141541-1A, wherein spatial component is at least partly through compressed version It is shown in bit stream using the field of the specified prediction mode used when compression space component.

Clause 141541-3A.The device of any combination of clause 141541-1A and clause 141541-2A, wherein space is divided Amount be through compressed version the huffman table at least partly used using specified when compression space component huffman table information It is shown in bit stream.

Clause 141541-4A.The device of any combination of clause 141541-1A to clause 141541-3A, wherein space is divided Amount is the quantization step or its variable at least partly used using instruction expression when compression space component through compressed version The field of value is shown in bit stream.

Clause 141541-5A.The device of clause 141541-4A, wherein described value includes nbits value.

Clause 141541-6A.The device of any combination of clause 141541-4A and clause 141541-5A, wherein the position Stream includes multiple spatial components of sound field (comprising its spatial component through compressed version) through compressed version, and wherein described value Express the quantization step or its variable used when compressing the multiple spatial component.

Clause 141541-7A.The device of any combination of clause 141541-1A to clause 141541-6A, wherein space is divided Amount is the classification that the compression classification for identifying that spatial component corresponds at least partly is indicated using Huffman code through compressed version Identifier is shown in bit stream.

Clause 141541-8A.The device of any combination of clause 141541-1A to clause 141541-7A, wherein space is divided Amount through compressed version be at least partly using identification spatial component be on the occasion of or the sign bit of negative value be shown in bit stream In.

Clause 141541-9A.The device of any combination of clause 141541-1A to clause 141541-8A, wherein space is divided Amount is at least partly to be shown in bit stream using Huffman code with the residual value of representation space component through compressed version.

Clause 141541-10A.The device of any combination of clause 141541-1A to clause 141541-9A, wherein described Device includes audio coding apparatus bit stream generation device.

Clause 141541-12A.The device of any combination of clause 141541-1A to clause 141541-11A, wherein being based on The synthesis of vector includes singular value decomposition.

Although being described as being executed by audio coding apparatus 510H, the technology can also by audio decoding apparatus 24 and/ Or any one of 540 execute.

By this method, in addition audio coding apparatus 510H can execute the various of the technology illustrated below in relation to following clause Aspect.

Clause 141541-1D.A kind of device, such as audio coding apparatus 510H comprising: one or more processors, institute One or more processors are stated to be configured to generate the bit stream through compressed version of the spatial component including sound field, the spatial component It is to be generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

Clause 141541-2D.The device of clause 141541-1D, wherein one or more described processors are further configured To generate the bit stream of the field of the prediction mode used comprising specified when compression space component when generating bit stream.

Clause 141541-3D.The device of any combination of clause 141541-1D and clause 141541-2D, wherein described one Or multiple processors are further configured to when generating bit stream, are generated comprising the specified Hough used when compression space component The huffman table bit stream of graceful table.

Clause 141541-4D.The device of any combination of clause 141541-1D to clause 141541-3D, wherein described one Or multiple processors are further configured to when generating bit stream, are generated comprising indicating what expression was used when compression space component The bit stream of the field of the value of quantization step or its variable.

Clause 141541-5D.The device of clause 141541-4D, wherein described value includes nbits value.

Clause 141541-6D.The device of any combination of clause 141541-4D and clause 141541-5D, wherein described one Or multiple processors are further configured to when generating bit stream, generate comprising sound field (comprising its spatial component through compressed version Originally the bit stream through compressed version of multiple spatial components), and wherein described value is expressed when compressing the multiple spatial component The quantization step used or its variable.

Clause 141541-7D.The device of any combination of clause 141541-1D to clause 141541-6D, wherein described one Or multiple processors are further configured to when generating bit stream, generate the compression class corresponded to comprising indicating identification spatial component The bit stream of the Huffman code of other class identifier.

Clause 141541-8D.The device of any combination of clause 141541-1D to clause 141541-7D, wherein described one Or multiple processors are further configured to when generating bit stream, generate comprising identification spatial component be on the occasion of or negative value just The bit stream of minus zone.

Clause 141541-9D.The device of any combination of clause 141541-1D to clause 141541-8D, wherein described one Or multiple processors are further configured to when generating bit stream, generate the Huffman code of the residual value comprising representation space component Bit stream.

Clause 141541-10D.The device of any combination of clause 141541-1D to clause 141541-10D, wherein being based on The synthesis of vector includes singular value decomposition.

Audio coding apparatus 510H can further be configured to implement the various sides of the technology as set forth in following clause Face.

Clause 141541-1E.A kind of device, such as audio coding apparatus 510H comprising: one or more processors, institute The spatial component that one or more processors are configured to compression sound field is stated, the spatial component is by relative to the humorous system of multiple balls Number executes the synthesis based on vector and generates.

Clause 141541-2E.The device of clause 141541-1E, wherein one or more described processors are further configured When compression space component, spatial component is transformed into integer representation from floating point representation.

Clause 141541-3E.The device of any combination of clause 141541-1E and clause 141541-2E, wherein described one Or multiple processors are further configured to be dynamically determined the value of instruction quantization step when compression space component, and are based on institute Value quantization spatial component is stated to generate quantified spatial component.

Clause 141541-4E.Such as the device of any combination of technical solution 1E to 3E, wherein one or more described processors It is further configured to when compression space component, the classification that identification spatial component corresponds to.

Clause 141541-5E.The device of any combination of clause 141541-1E and clause 141541-4E, wherein described one Or multiple processors are further configured to identify the residual value of the spatial component when compression space component.

Clause 141541-6E.The device of any combination of clause 141541-1E to clause 141541-5E, wherein described one Or multiple processors are further configured to when compression space component, are executed relative to spatial component and subsequent spatial component Prediction is to generate predicted spatial component.

Clause 141541-7E.The device of any combination of clause 141541-1E, wherein one or more described processors into One step is configured to when compression space component, and spatial component is transformed into integer representation from floating point representation, is dynamically determined instruction The value of quantization step, and based on described value quantization spatial component integer representation to generate quantified spatial component, based on through measuring Change spatial component and identifies that the classification that the spatial component corresponds to generate class identifier, determines the positive and negative of the spatial component Number, identify the residual value of the spatial component based on quantified spatial component and class identifier, and based on class identifier, just Negative sign and residual value generate spatial component through compressed version.

Clause 141541-8E.The device of any combination of clause 141541-1E, wherein one or more described processors into One step is configured to when compression space component, and spatial component is transformed into integer representation from floating point representation, is dynamically determined instruction The value of quantization step, the integer representation based on described value quantization spatial component to generate quantified spatial component, execute relative to The prediction of spatial component and subsequent spatial component is identified predicted with generating predicted spatial component based on quantified spatial component The classification that spatial component corresponds to determines the sign of the spatial component to generate class identifier, is based on quantified space Component and class identifier identify the residual value of the spatial component, and are generated based on class identifier, sign and residual value Spatial component through compressed version.

Clause 141541-9E.The device of any combination of clause 141541-1E to clause 141541-8E, wherein based on to The synthesis of amount includes singular value decomposition.

Furthermore the various aspects of the technology can enable audio coding apparatus 510H be configured to such as in following clause It is illustrated and operate.

Clause 141541-1F.A kind of device, such as audio coding apparatus 510H comprising: one or more processors, institute One or more processors are stated to be configured to based on current spatial component in multiple spatial components relative to the multiple spatial component In the rest order identification huffman codebook to be used when compressing the current spatial component, the spatial component is logical It crosses and executes the synthesis based on vector relative to multiple spherical harmonic coefficients and generate.

Clause 141541-2F.The device of clause 141541-3F, wherein one or more described processors are further configured With implementation of the provisions 141541-1A to clause 141541-12A, clause 141541-1B to clause 141541-10B and clause Any combination for the step of being described in 141541-1C to clause 141541-9C.

Furthermore the various aspects of the technology can enable audio coding apparatus 510H be configured to such as in following clause It is illustrated and operate.

Clause 141541-1H.A kind of device, such as audio coding apparatus 510H comprising: one or more processors, institute It states one or more processors and is configured to determine the quantization step used when the spatial component for staying in compression sound field, the space point Amount is generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

Clause 141541-2H.The device of clause 141541-1H, wherein one or more described processors are further configured To determine quantization step based on targeted bit rates when determining quantization step.

Clause 141541-3H.The device of clause 141541-1H, wherein one or more described processors are further configured When selecting one of multiple quantization steps, to determine the estimation of the number of the position to representation space component, and based on estimating Difference between meter and targeted bit rates determines quantization step.

Clause 141541-4H.The device of clause 141541-1H, wherein one or more described processors are further configured To determine the estimation of the number of the position to representation space component when selecting one of multiple quantization steps, estimation is determined Difference between targeted bit rates, and by the way that the difference is determined quantization step with targeted bit rates phase Calais.

Clause 141541-5H.The device of clause 141541-3H or clause 141541-4H, wherein one or more described processing Device be further configured to calculate when determining the estimation of bits number the bits number to be generated for spatial component estimation (to Surely correspond to the codebook of targeted bit rates).

Clause 141541-6H.The device of clause 141541-3H or clause 141541-4H, wherein one or more described processing Device be further configured to calculate when determining the estimation of bits number the bits number to be generated for spatial component estimation (to The decoding mode used when settled compression space component).

Clause 141541-7H.The device of clause 141541-3H or clause 141541-4H, wherein one or more described processing Device is further configured to calculate the first estimation of the bits number to generate for spatial component when determining the estimation of bits number (given the first decoding mode used when staying in compression space component) calculates the of the bits number to generate for spatial component Two estimations (give the second decoding mode used when staying in compression space component), select to have in the first estimation and the second estimation Wait be used as one of the minimum number of bits through determining estimation of bits number.

Clause 141541-8H.The device of clause 141541-3H or clause 141541-4H, wherein one or more described processing Device is further configured to when determining the estimation of bits number, the classification identification for the classification that identification identification spatial component corresponds to Symbol identifies the bit length of the residual value for being used for spatial component of generation when compression corresponds to the spatial component of classification, and leads to Cross the estimation that bits number will be at least partly determined to indicate the bits number of class identifier with the bit length phase Calais of residual value.

Clause 141541-9H.The device of any combination of clause 141541-1H to clause 141541-8H, wherein based on to The synthesis of amount includes singular value decomposition.

Although being described as being executed by audio coding apparatus 510H, the above clause clause 141541-1H to clause 141541- The technology illustrated in 9H can also be executed by audio decoding apparatus 540D.

In addition, the various aspects of the technology can enable audio coding apparatus 510H be configured to such as in following clause It is illustrated and operate.

Clause 141541-1J.A kind of device, such as audio coding apparatus 510J comprising: one or more processors, institute It states one or more processors and is configured to one of the multiple codebooks used when selecting to stay in the spatial component of compression sound field, institute Stating spatial component is generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

Clause 141541-2J.The device of clause 141541-1J, wherein one or more described processors are further configured To determine the position to representation space component using each of the multiple codebook when selecting one of multiple codebooks Cause with minimum number of bits purpose in the estimation of number, and the multiple codebook of selection through determining one of estimation.

Clause 141541-3J.The device of clause 141541-1J, wherein one or more described processors are further configured To determine the digit to representation space component using one or more of multiple codebooks when selecting one of multiple codebooks Purpose estimates that the one or more in the multiple codebook is the element based on spatial component to be compressed relative to space point Amount other order of element numbers and select.

Clause 141541-4J.The device of clause 141541-1J, wherein one or more described processors are further configured With when selecting one of multiple codebooks, using being designed to make when not predicting the spatial component from subsequent spatial component One of multiple codebooks determine the estimation of the bits number to representation space component.

Clause 141541-5J.The device of clause 141541-1J, wherein one or more described processors are further configured With when selecting one of multiple codebooks, using being designed to the use when predicting the spatial component from subsequent spatial component One of multiple codebooks determine the estimation of the bits number to representation space component.

Clause 141541-6J.The device of clause 141541-1J, wherein one or more described processors are further configured With when selecting one of multiple codebooks, using being designed to make when spatial component indicates the Composite tone object in sound field One of multiple codebooks determine the estimation of the bits number to representation space component.

Clause 141541-7J.The device of clause 141541-1J, wherein the Composite tone object includes through pulse-code modulation (PCM) audio object.

Clause 141541-8J.The device of clause 141541-1J, wherein one or more described processors are further configured With when selecting one of multiple codebooks, using being designed to when spatial component indicates the recorded audio object in sound field One of multiple codebooks used determine the estimation of the bits number to representation space component.

Clause 141541-9J.The device of any combination of technical solution 1J to 8J, wherein the synthesis based on vector includes surprise Different value is decomposed.

In each of above-mentioned various examples, it should be appreciated that 510 executing method of audio coding apparatus, or in addition packet Include the device for executing each step that audio coding apparatus 510 is configured to the method executed.In some cases, these devices It may include one or more processors.In some cases, one or more described processors can be indicated by means of storage to nonvolatile Property computer-readable storage medium instruction configuration application specific processor.In other words, each of the set of encoding example In the various aspects of technology can provide the non-transitory computer-readable storage media for being stored thereon with instruction, described instruction exists One or more described processors are caused to execute the method that audio coding apparatus 510 has been configured to execute when execution.

Figure 40 J is that the various aspects of technology described in the present invention can be performed with packed description two dimension or three dimensional sound in explanation The block diagram of the example audio code device 510J of the spherical harmonic coefficient of field.Audio coding apparatus 510J can be similar to audio coding apparatus 510G, because audio coding apparatus 510J includes audio compressing unit 512, audio coding unit 514 and bitstream producing unit 516.In addition, the audio compressing unit 512 of audio coding apparatus 510J can be similar to the audio compression of audio coding apparatus 510G Unit, because audio compressing unit 512 includes decomposition unit 518 and sound field component extraction unit 520, the unit be can be similar to The same unit of audio coding apparatus 510I operates.In some instances, audio coding apparatus 510J may include quantifying unit 534 (as relative to described by Figure 40 D to 40E) is to quantify UDISTVector 525C, UBGVector 525D, VT DISTVector 525E and VT BGVector One or more vectors of any one of 525J.

However, the audio compressing unit 512 of audio coding apparatus 510J may differ from the audio of audio coding apparatus 510G Compression unit 512, the difference is that, the audio compressing unit 512 of audio coding apparatus 510J includes to be expressed as interpolation list The additional unit of member 550.Interpolation unit 550 can indicate the first audio frame of subframe of the interpolation from the first audio frame subframe and Second time upper subsequent or previous audio frame, is such as more fully described below in relation to Figure 45 and 45B.Interpolation unit 550 can It is calculated again by potentially reducing the degree for needing decomposition unit 518 to decompose SHC 511 to reduce in executing this interpolation process Miscellaneous degree (for processing cycle and/or memory consumption).Interpolation unit 550 can by being similar to above in regard in Figure 40 H and The mode for the mode that the interpolation unit 550 of audio coding apparatus 510H and 510I shown in the example of 40I describe operates.

In operation, interpolation unit 200 can interpolation from one comprising a spherical harmonic coefficient 11 more than first in the first frame Partial first decomposes the of (for example, V matrix 19') and a part comprising a spherical harmonic coefficient 11 more than second in the second frame Two decompose one or more subframes of first audio frame of (for example, V matrix 19') to generate the warp for one or more subframes The interpolated spherical harmonic coefficient decomposed.

Interpolation unit 550 can decompose and more than second at least partially through first relative to more than first a spherical harmonic coefficients The second of spherical harmonic coefficient decomposes the interpolated spherical harmonic coefficient through decomposing for executing interpolation and obtaining a time slice.Smooth unit 554 can be using the interpolated spherical harmonic coefficient through decomposing so that a spherical harmonic coefficient more than described first and more than second a spherical harmonic coefficient Spatial component and at least one of time component it is smooth.Smooth unit 554 can produce smoothed UDISTMatrix 525C', As above relative to described by Figure 37 to 39.First and second decomposition can refer to the V in Figure 40 J1 T 556、V2 T 556B。

In some cases, VTOr other V vectors or V matrix can be exported with quantified version to be used for interpolation.With this side Formula, the V vector for interpolation can be identical as the V vector at decoder, and V vector interpolation also can be performed in decoder, such as to restore Multidimensional signal.

In some instances, first decompose include indicate the right side of the part of more than first a spherical harmonic coefficient 511 it is unusual to First V matrix 519' of amount.Similarly, in some instances, it second decomposes comprising indicating more than second a spherical harmonic coefficient 2nd V matrix 519' of partial right singular vector.

Interpolation unit 550 can be held based on the first V matrix 519' and the 2nd V matrix 19' relative to one or more described subframes Row temporal interpolation.That is, interpolation unit 550 can be based on the V matrix 519' decomposed from the first subframe of the first audio frame and from second The V matrix 519' that first subframe of audio frame is decomposed four whole subframes of the interpolation for example from the first audio frame in time Second, third and the 4th subframe.In some instances, it is inserted as linear time interpolation in this time, wherein when the first audio of interpolation Fourth son of the V matrix 519' ratio decomposed when the second subframe of frame from the first subframe of the first audio frame when the first audio frame of interpolation Weight heavier when frame.When interpolation third subframe, V matrix 519' can be equably weighted.When four subframe of interpolation, from second The V matrix 519' that first subframe of audio frame is decomposed is comparable from the important place V matrix 519' that the first subframe of the first audio frame is decomposed Weighting.

In other words, the proximity of one of subframe of the first audio frame to be interpolated is given, then linear time interpolation V matrix 519' can be weighted.For the second subframe to be interpolated, the given V matrix decomposed from the first subframe of the first audio frame The proximity of 519' and the second subframe to be interpolated can then compare the first subframe from the second audio frame to the V matrix 519' The important place matrix 519' V of decomposition weights.For this reason, when being based on V matrix 519' interpolation third subframe, weight can be equivalent. The first subframe of the 4th subframe to be interpolated closer second audio frame compared with the first subframe close to the first audio frame is given, The weight for being then applied to the V matrix 519' decomposed from the first subframe of the second audio frame, which can be greater than, to be applied to from the first audio frame The weight for the V matrix 519' that first subframe is decomposed.

In some instances, the first V matrix 519' that interpolation unit 550 can will be decomposed from the first subframe of the first audio frame It projects in spatial domain to generate first through Projective decomposition.In some instances, this projection covers the projection (example in spherical surface Such as, using projection matrix, such as T design matrix).Interpolation unit 550 will then can be decomposed from the first subframe of the second audio frame The 2nd V matrix 519' project in spatial domain to generate second through Projective decomposition.Interpolation unit 550 can attaching space interpolation (it can be linear interpolation again) first generated through Projective decomposition and second through Projective decomposition the first spatial interpolation through projecting Decompose and second space interpolation through Projective decomposition.Interpolation unit 550 can be then based on the first spatial interpolation through Projective decomposition And second space interpolation through one or more subframes described in Projective decomposition in time interpolation.

In 550 space of interpolation unit and then those of time projection V matrix 519' example, interpolation unit 550 can be incited somebody to action The temporal interpolation spherical harmonic coefficient generated from one or more subframes described in interpolation projects back to the humorous domain of ball, and then generates V matrix 519, S Matrix 519B and U matrix 519C.

In some instances, the part of a spherical harmonic coefficient more than described first includes more than described first a spherical harmonic coefficients 511 Single subframe.In some instances, the part of a spherical harmonic coefficient more than described second includes a spherical harmonic coefficient 511 more than described second Single subframe.In some instances, it is the first subframe that V matrix 19', which decomposes this single subframe certainly,.

In some instances, first frame is divided into four subframes.In these and other example, a ball more than described first The part of humorous coefficient only includes the first subframe of the multiple spherical harmonic coefficient 511.In these and other example, the second frame is through drawing It is divided into four subframes, and the part of a spherical harmonic coefficient 511 more than described second only includes of a spherical harmonic coefficient 511 more than described second One subframe.

Although in some instances, executing interpolation, a ball more than described first using only the first subframe of each audio frame The part of humorous coefficient may include two in four subframes of a spherical harmonic coefficient 511 more than described first.In these and other example In, the part of a spherical harmonic coefficient 511 more than described second includes two in four subframes of a spherical harmonic coefficient 511 more than described second It is a.

As noted before, single device (for example, audio coding apparatus 510J) can be also to decompose a ball more than described first humorous Interpolation is executed while the part of coefficient to generate the first decomposition of the part of a spherical harmonic coefficient more than described first.At these And in other examples, decomposition unit 518 can decompose the part of a spherical harmonic coefficient more than described second to generate a ball more than described second The second of the part of humorous coefficient is decomposed.Although describing relative to single device, this hair is can be performed in two or more devices Technology described in bright, wherein one of described two devices execute decomposition, and the other of described device is according to this Technology described in invention executes interpolation.

In some instances, decomposition unit 518 can execute unusual relative to the part of more than described first a spherical harmonic coefficients 511 Value is decomposed to generate the V matrix 519'(and s-matrix 519B' that indicate the right singular vector of a spherical harmonic coefficient 511 more than described first And U matrix 519C', purpose for ease of explanation its do not show).In these and other example, decomposition unit 518 can be opposite Executing singular value decomposition in the part of more than described second a spherical harmonic coefficients 511 indicates a spherical harmonic coefficient more than described second to generate The V matrix 519'(and s-matrix 519B' and U matrix 519C' of 511 right singular vector, purpose for ease of explanation its not It shows).

In some instances, as noted before, first and second described multiple spherical harmonic coefficient respectively indicate the plane of sound field Wave table shows.In these and other example, first and second described multiple spherical harmonic coefficient 511 respectively indicate mix one Or multiple monophonic audio objects.

It in other words, can be the ginseng of 3D pressure field based on the humorous 3D audio of ball for the orthogonal basement function on spherical surface Number indicates.The order N of the expression is higher, and spatial resolution is possibly higher, and the number of usually humorous (SH) coefficient of ball is bigger ((N+1) in total2A coefficient).For many applications, it may be necessary to which the bandwidth reduction of coefficient can be transmitted effectively and store institute State coefficient.This targeted in the present invention technology can provide to be reduced using the dimension based on frame of singular value decomposition (SVD) Journey.Each frame of coefficient can be resolved into three matrix Us, S and V by SVD analysis.In some instances, the technology can will be in U Some durection component disposition as basic sound field in vector.However, when disposing by this method, these vectors (in U) Even if being discontinuous in interframe -- it indicates same different audio component.When passing through component described in the feed-in of converting audio frequency decoder When, these discontinuities can lead to significant illusion.

Technology described in the present invention can solve this discontinuity.That is, the technology can be based on following observed result: V Matrix can the interpreted orthogonal intersection space axis in the humorous domain of ball.U matrix can indicate humorous (HOA) data of ball according to those basis functions Projection, wherein discontinuity is attributable to basis function (V), and the basis function (V) changes each frame and therefore itself is not Continuously.This is different from the similar decomposition of such as Fourier transform, and wherein basis function will be in some instances normal in interframe Number.In these terms, SVD is regarded as matching and pursues algorithm.Technology described in the present invention can enable interpolation unit 550 It is enough to maintain the continuity between basis function (V) in interframe and progress interpolation therebetween.

In some instances, the technology enables interpolation unit 550 that the frame of SH data is divided into four subframes, as above It is described and further described below in relation to Figure 45 and 45B.Interpolation unit 550 can then calculate the SVD of the first subframe.It is similar Ground, we calculate the SVD of the first subframe of the second frame.For each of first frame and the second frame, interpolation unit 550 can lead to Space reflection will be transformed into for the vector in V (using the projection matrix of such as T design matrix) on vector projection to spherical surface by crossing.It is interior Can then the vector in V be interpreted as the shape on spherical surface by inserting unit 550.In order in the first subframe of first frame and next frame The first subframe between three subframes of interpolation V matrix, interpolation unit 550 can then these spatial forms of interpolation, and then pass through Inverse by projection matrix is transformed back to SH vector.By this method, technology of the invention can provide smooth turn between V matrix Become.

Figure 41 to 41D be respective explanation can be performed the various aspects of technology described in the present invention with to description two dimension or The block diagram for example audio decoding device 540A to the 540D that the spherical harmonic coefficient of three-dimensional sound field is decoded.Audio decoding apparatus 540A It can indicate any device for capableing of decoding audio data, such as desktop PC, laptop computer, work station, plate meter It is calculation machine, special audio recording device, cellular phone (include so-called " smart phone "), personal media player device, a People's game device or be capable of decoding audio data any other type device.

In some instances, audio decoding apparatus 540A is executed and by any one of audio coding apparatus 510 or 510B The reciprocal audio decoding process of the audio encoding process of execution, exception are to execute Degree reduction (as above to arrive relative to Figure 40 B Described by the example of 40J), in some instances, the audio decoding process is by audio coding apparatus 510B to 510J to have Help the removal of external uncorrelated data.

Although being shown as single device, that is, the device 540A in the example of Figure 41, mentioned below is to be contained in device Various assemblies or unit in 540A may be formed at the isolated system outside device 540.In other words, although being described in the present invention To be executed by single device (that is, device 540A in the example of Figure 41), but the technology can be by being including multiple devices System implement or in addition execute, wherein each of these devices can respectively contain the various assemblies being described more particularly below or One or more of unit.Therefore, the example that the technology should not necessarily be limited by Figure 41 in this respect.

As shown in the example of Figure 41, audio decoding apparatus 540A includes extraction unit 542, audio decoding unit 544, number Learn unit 546 and audio reproduction unit 548.The expression of extraction unit 542 is configured to extract encoded reduction background from bit stream 517 Spherical harmonic coefficient 515B, encoded UDIST*SDISTVector 515A and VT DISTThe unit of vector 525E.Extraction unit 542 will be encoded Reduce background spherical harmonic coefficient 515B and encoded UDIST*SDISTVector 515A is output to audio decoding unit 544, while also will VT DISTMatrix 525E is output to mathematical units 546.In in this respect, extraction unit 542 can be similar to shown in the example of Fig. 5 The mode of extraction unit 72 of audio decoding apparatus 24 operate.

Audio decoding unit 544 indicate to decode coded audio data (often according to reciprocal audio decoding scheme, such as AAC decoding scheme) to restore UDIST*SDISTVector 527 and the unit for reducing background spherical harmonic coefficient 529.Audio decoding unit 544 by UDIST*SDISTVector 527 and reduced background spherical harmonic coefficient 529 are output to mathematical units 546.In in this respect, audio Decoding unit 544 can be by the psychologic acoustics decoding unit 80 for being similar to the audio decoding apparatus 24 shown in the example of fig. 5 Mode operates.

Mathematical units 546 can indicate to be configured to execute matrix multiplication and addition (and in some instances, Ren Heqi Its matrix mathematical operation) unit.U can be first carried out in mathematical units 546DIST*SDISTVector 527 and VT DISTThe square of matrix 525E Battle array multiplication.Mathematical units 546 can be then by UDIST*SDISTVector 527 and VT DISTMatrix 525E and reduced background spherical harmonic coefficient The result of 529 multiplication (again, can be referred to UBGMatrix 525D and SBGMatrix 525B and then with VT BGThe multiplication of matrix 525F Result) and UDIST*SDISTVector 527 and VT DISTThe results added of the matrix multiplication of matrix 525E is to generate original spherical harmonic coefficient 11 reduction version (it is expressed as the spherical harmonic coefficient 547 restored).Mathematical units 546 can export the spherical harmonic coefficient 547 of recovery To audio reproduction unit 548.In in this respect, mathematical units 546 can be by similar to the audio decoder shown in the example of fig. 5 The prospect of device 24 works out unit 78 and the mode of HOA coefficient formulation unit 82 operates.

The expression of audio reproduction unit 548 is configured to reproduction channel 549A to 549N, and (" channel 549 ", can also be substantially Referred to as " multichannel audio data 549 " or make " speaker feeds 549 ") unit.Audio reproduction unit 548 can will convert (often expressing by the form of matrix) is applied to the spherical harmonic coefficient 547 restored.Because the spherical harmonic coefficient 547 restored describes in three-dimensional Sound field, so the expression of spherical harmonic coefficient 547 restored facilitates can adapt to most decoders-part loudspeaker geometrical arrangements The mode of (it can refer to that the geometrical arrangements of the loudspeaker of multichannel audio data 549 will be reset) reproduces multichannel audio data 549A Audio format.More information above in regard to Figure 48 description about the reproduction of multichannel audio data 549A.

Although being described in the situation that multichannel audio data 549A is surround sound multichannel audio data 549, audio Reproduction unit 48 can also carry out the form of alliteration so that restore spherical harmonic coefficient 549A alliteration, and in turn generate two alliterations The channel 549 of reproduction.Therefore, the technology should not necessarily be limited by the surround sound form of multichannel audio data, and may include alliteration Multichannel audio data.

The various aspects of technology described in the present invention can be presented in the various clauses being listed below.

Clause 132567-1B.A kind of device, such as audio decoding apparatus 540 comprising: one or more processors, it is described One or more processors are configured to determine one or more primary vectors of the distinct components of description sound field and describe the back of sound field One or more secondary vectors of scape component, both one or more described primary vectors and one or more secondary vectors are at least logical It crosses and executes the singular value decomposition relative to the multiple spherical harmonic coefficient and generate.

Clause 132567-2B.The device of clause 132567-1B, wherein one or more described primary vectors are included in audio By by one or more U through audio coding of U matrix before codingDISTVector multiplied by s-matrix one or more SDISTVector and produce U of one or more raw through audio codingDIST*SDISTVector, wherein U matrix and s-matrix are at least through executing relative to described more The singular value decomposition of a spherical harmonic coefficient and generate, and wherein one or more described processors are further configured to audio decoder institute State one or more U through audio codingDIST*SDISTVector is to generate one or more described U through audio codingDIST*SDISTVector Through audio decoder version.

Clause 132567-3B.The device of clause 132567-1B, wherein one or more described primary vectors are included in audio By by one or more U through audio coding of U matrix before codingDISTVector multiplied by s-matrix one or more SDISTVector and V One or more V of the transposition of matrixT DISTVector and one or more U through audio coding generatedDIST*SDISTVector, wherein U square Battle array and s-matrix and V matrix are generated at least through the singular value decomposition relative to the multiple spherical harmonic coefficient is executed, and wherein institute It states one or more processors and is further configured to one or more U through audio coding described in audio decoderDIST*SDISTVector with Generate one or more described U through audio codingDIST*SDISTVector through audio decoder version.

Clause 132567-4B.The device of clause 132567-3B, wherein one or more described processors are further configured With by UDIST*SDISTVector is multiplied by VT DISTVector is to restore those of the humorous middle distinct components for indicating sound field of the multiple ball ball It is humorous.

Clause 132567-5B.The device of clause 132567-1B, wherein one or more described secondary vectors are included in audio By that will include the U in U matrix before codingBGVector is multiplied by including S in s-matrixBGVector and then multiplied by be included in V V in the transposition of matrixT BGVector and one or more U through audio coding generatedBG*SBG*VT BGVector, wherein s-matrix, U square Battle array and V matrix are respectively generated at least through the singular value decomposition relative to the multiple spherical harmonic coefficient is executed.

Clause 132567-6B.The device of clause 132567-1B, wherein one or more described secondary vectors are included in audio By that will include the U in U matrix before codingBGVector is multiplied by including S in s-matrixBGVector and then multiplied by be included in V V in the transposition of matrixT BGVector and one or more U through audio coding generatedBG*SBG*VT BGVector, and wherein s-matrix, U Matrix and V matrix are generated at least through the singular value decomposition relative to the multiple spherical harmonic coefficient is executed, and wherein described one Or multiple processors are further configured to one or more U through audio coding described in audio decoderBG*SBG*VT BGVector is to produce One or more raw U through audio decoderBG*SBG*VT BGVector.

Clause 132567-7B.The device of clause 132567-1B, wherein one or more described primary vectors are included in audio By by one or more U through audio coding of U matrix before codingDISTVector multiplied by s-matrix one or more SDISTVector and V One or more V of the transposition of matrixT DISTVector and one or more U through audio coding generatedDIST*SDISTVector, wherein U square Battle array, s-matrix and V matrix are generated at least through the singular value decomposition relative to the multiple spherical harmonic coefficient is executed, and wherein institute It states one or more processors and is further configured to one or more U through audio coding described in audio decoderDIST*SDISTVector with Generate one or more UDIST*SDISTVector, and by UDIST*SDISTVector is multiplied by VT DISTVector is to restore the multiple spherical harmonic coefficient Those of the distinct components of middle description sound field spherical harmonic coefficient, wherein one or more described secondary vectors are included in front of audio coding just Pass through the U being included in U matrixBGVector is multiplied by the S being contained in s-matrixBGVector and then multiplied by being contained in V matrix V in transpositionT BGVector and one or more U through audio coding generatedBG*SBG*VT BGVector, and wherein it is described one or more Processor is further configured to one or more U through audio coding described in audio decoderBG*SBG*VT BGVector is to restore to describe At least part of the multiple spherical harmonic coefficient of the background component of sound field, and will describe sound field distinct components it is the multiple Spherical harmonic coefficient is added the multiple to generate at least part of the multiple spherical harmonic coefficient of the background component of description sound field The reconstructed version of spherical harmonic coefficient.

Clause 132567-8B.The device of clause 132567-1B, wherein one or more described primary vectors are included in audio By by one or more U through audio coding of U matrix before codingDISTVector multiplied by s-matrix one or more SDISTVector and V One or more V of the transposition of matrixT DISTVector and one or more U generatedDIST*SDISTVector, wherein U matrix, s-matrix and V Matrix is generated at least through the singular value decomposition relative to the multiple spherical harmonic coefficient is executed, and wherein described at one or more Reason device is further configured to determine that instruction will be extracted from bit stream to form one or more described UDIST*SDISTOne described in vector sum Or multiple VT DISTThe value D of the number of vectors of vector.

Clause 132567-9B.The device of clause 132567-10B, wherein one or more described primary vectors are included in audio By by one or more U through audio coding of U matrix before codingDISTVector multiplied by s-matrix one or more SDISTVector and V One or more V of the transposition of matrixT DISTVector and one or more U generatedDIST*SDISTVector, wherein U matrix, s-matrix and V Matrix is generated at least through singular value decomposition is executed relative to the multiple spherical harmonic coefficient, and wherein described at one or more Reason device be further configured to based on audio frame one by one determine value D, described value D indicate to from bit stream extraction with formed described one or Multiple UDIST*SDISTVector and one or more described VT DISTThe number of the vector of vector.

Clause 132567-1G.A kind of device, such as audio decoding apparatus 540 comprising: one or more processors, it is described Or multiple processors are configured to determine one or more primary vectors of the distinct components of description sound field and describe the background of sound field One or more secondary vectors of component, both one or more described primary vectors and one or more secondary vectors at least through It executes the singular value decomposition relative at least part of multichannel audio data for indicating sound field and generates.

Clause 132567-2G.The device of clause 132567-1G, wherein the multichannel audio data includes that multiple balls are humorous Coefficient.

Clause 132567-3G.The device of clause 132567-2G, wherein one or more described processors are further configured With implementation of the provisions 132567-2B to any combination of clause 132567-9B.

From each of above-mentioned various clauses, it should be appreciated that any one of audio decoding apparatus 540A to 540D can be held Row method, or also comprise the dress for executing each step that audio decoding apparatus 540A to 540D is configured to the method executed It sets.In some cases, these devices may include one or more processors.In some cases, one or more described processors It can indicate by means of storing the application specific processor to the instruction configuration of non-transitory computer-readable storage media.In other words, The various aspects of technology in each of set of encoding example can provide the non-transitory calculating for being stored thereon with instruction Machine readable memory medium, described instruction cause one or more processors execution audio decoding apparatus 540A to arrive when being executed 540D has been configured to the method executed.

For example, clause 132567-10B can export as method comprising the following steps from aforementioned clause 132567-1B: Determine description sound field distinct components one or more primary vectors and describe sound field background component one or more second to Amount, both one or more described primary vectors and one or more secondary vectors are at least through execution relative to expression sound field The singular value decomposition of multiple spherical harmonic coefficients and generate.

As another example, clause 132567-11B can export as a kind of device, such as sound from aforementioned clause 132567-1B Frequency decoding apparatus 540 comprising for determining one or more primary vectors of the distinct components of description sound field and describing sound field The device of one or more secondary vectors of background component, one or more described primary vectors and one or more described secondary vectors two Person generates at least through the singular value decomposition relative to the multiple spherical harmonic coefficient is executed;And for store it is described one or more The device of primary vector and one or more secondary vectors.

As yet another embodiment, clause 132567-12B can be exported as from clause 132567-1B it is a kind of have be stored thereon Instruction non-transitory computer-readable storage media, described instruction cause when being executed one or more processors determine description One or more secondary vectors of the background component of one or more primary vectors and description sound field of the distinct components of sound field, described one Or both multiple primary vectors and one or more secondary vectors at least through execute relative to be contained in description sound field compared with The singular value decomposition of multiple spherical harmonic coefficients in high-order ambiophony audio data and generate.

For as illustrated and derived various devices, method and the computer-readable storage matchmaker of non-transitory above Body, various clauses can be exported similarly from clause 132567-2B to 132567-9B.It can be listed for through the present invention various Other clauses execute same operation.

Figure 41 B is that the various aspects of technology described in the present invention can be performed with packed description two dimension or three dimensional sound in explanation The block diagram of the example audio decoding device 540B of the spherical harmonic coefficient of field.Audio decoding apparatus 540B can be similar to audio decoding apparatus 540, exception is that in some instances, extraction unit 542 can extract the V of rearranged sequenceT DISTVector 539, rather than VT DISTVector 525E.In other examples, extraction unit 542 can extract VT DISTVector 525E, and be then based on referring in bit stream Determine or is inferred (by the analysis of other vectors) with the V of the rearranged sequence of determinationT DISTThe information that reorders of vector 539 reorders this A little VT DISTVector 525E.In in this respect, extraction unit 542 can be by audio decoding apparatus shown in the example similar to Fig. 5 The mode of 24 extraction unit 72 operates.Under any circumstance, extraction unit 542 can be by the V of rearranged sequenceT DISTVector 539 is defeated Mathematical units 546 are arrived out, wherein can be relative to the V of these rearranged sequencesT DISTVector 539 is executed above in regard to the recovery humorous system of ball The process of number description.

By this method, the technology can enable the distinct components of audio decoding apparatus 540B audio decoder expression sound field One or more vectors of rearranged sequence, one or more vectors of the rearranged sequence have reordered to help to compress described one Or multiple vectors.In these and other example, audio decoding apparatus 540B can be by one or more vectors of rearranged sequence and warp The additional vector of one or more to reorder is recombinated to restore to indicate the spherical harmonic coefficient of the distinct components of sound field.In these and other reality In example, audio decoding apparatus 540B can the spherical harmonic coefficient then based on the distinct components for indicating sound field and the back based on expression sound field The spherical harmonic coefficient of scape component restores multiple spherical harmonic coefficients.

That is, the various aspects of the technology can provide wait be configured to decode the one or more of rearranged sequence according to the following terms The audio decoding apparatus 540B of a vector.

Clause 133146-1F.A kind of device, such as audio coding apparatus 540B comprising: one or more processors, institute State the number for the vector that one or more processors are configured to determine corresponding to the component in sound field.

Clause 133146-2F.The device of clause 133146-1F, wherein one or more described processors are configured to root The number of determining vector after Degree reduction is executed according to any combination of examples detailed above.

Clause 133146-3F.The device of clause 133146-1F, wherein one or more described processors are further configured To execute Degree reduction according to any combination of examples detailed above.

Clause 133146-4F.The device of clause 133146-1F, wherein one or more described processors are configured to from position The value specified in stream determines the number of vector, and wherein one or more described processors are further configured to based on determined by Number of vectors dissects bit stream to identify one or more vectors for indicating the distinct components of sound field in bit stream.

Clause 133146-5F.The device of clause 133146-1F, wherein one or more described processors are configured to from position The value specified in stream determines the number of vector, and wherein one or more described processors are further configured to based on determined by Number of vectors dissects bit stream to identify one or more vectors for the background component for indicating sound field in bit stream.

Clause 133143-1C.A kind of device, such as audio decoding apparatus 540B comprising: one or more processors, institute State one or more processors be configured to reorder the distinct components for indicating sound field rearranged sequence one or more vectors.

Clause 133143-2C.The device of clause 133143-1C, wherein one or more described processors are further configured With one or more vectors of the rearranged sequence of determination, and determine the mode of one or more vectors of the rearranged sequence that describes to reorder The information that reorders, wherein one or more described processors are further configured to as the one or more of the rearranged sequence that reorder When a vector, reordered one or more vectors of the rearranged sequence based on the identified information that reorders.

Clause 133143-3C.The device of 1C, wherein one or more vectors of the rearranged sequence include by technical solution 1A To 18A any combination or technical solution 1B to 19B any combination narration described in one or more rearranged sequences first to Amount, and wherein one or more described primary vectors be according to by technical solution 1A to 18A any combination or technical solution 1B to The method of any combination narration of 19B determines.

Clause 133143-4D.A kind of device, such as audio decoding apparatus 540B comprising: one or more processors, institute State one or more processors be configured to audio decoder indicate sound field distinct components rearranged sequence one or more vectors, institute One or more vectors for stating rearranged sequence have reordered to help to compress one or more described vectors.

Clause 133143-5D.The device of clause 133143-4D, wherein one or more described processors are further configured To recombinate one or more vectors of the rearranged sequence and one or more additional vectors of rearranged sequence to restore to indicate sound field The spherical harmonic coefficient of distinct components.

Clause 133143-6D.The device of clause 133143-5D, wherein one or more described processors are further configured It is humorous to restore multiple balls with the spherical harmonic coefficient based on the distinct components for indicating sound field and the spherical harmonic coefficient for the background component for indicating sound field Coefficient.

Clause 133143-1E.A kind of device, such as audio decoding apparatus 540B comprising: one or more processors, institute One or more processors are stated to be configured to reorder one or more vectors to generate one or more primary vectors of rearranged sequence, and And then facilitate the coding carried out by old edition audio coder, wherein one or more described vector descriptions indicate different point of sound field Amount, and using one or more vectors of the rearranged sequence of old edition audio coder audio coding with generate rearranged sequence one or more The encoded version of vector.

Clause 133143-2E.The device of 1E, wherein one or more vectors of the rearranged sequence include by technical solution 1A To 18A any combination or technical solution 1B to 19B any combination narration described in one or more rearranged sequences first to Amount, and wherein one or more described primary vectors be according to by technical solution 1A to 18A any combination or technical solution 1B to The method of any combination narration of 19B determines.

Figure 41 C is the block diagram for illustrating another exemplary audio code device 540C.Audio decoding apparatus 540C can indicate energy Any device of enough decoding audio datas, for example, it is desktop PC, laptop computer, work station, tablet computer, dedicated Audio recording device, cellular phone (including so-called " smart phone "), personal media player apparatus, personal gaming device Or it is capable of the device of any other type of decoding audio data.

In the example of Figure 41 C, audio decoding apparatus 540C was executed and by being appointed in audio coding apparatus 510B to 510J The reciprocal audio decoding process of the audio encoding process that one executes, exception are to execute order to reduce (as above relative to figure Described by the example of 40B to 40J), in some instances, the audio decoding process is used by audio coding apparatus 510B to 510J To facilitate the removal of external uncorrelated data.

Although being shown as single device, that is, the device 540C in the example of Figure 41 C, mentioned below is to be contained in dress Setting various assemblies or unit in 540C may be formed at isolated system outside device 540C.In other words, although in the present invention It is described as being executed by single device (that is, device 540C in the example of Figure 41 C), but the technology can be by including multiple devices System implement or in addition execute, wherein each of these devices can respectively contain various groups be described more particularly below One or more of part or unit.Therefore, the example that the technology should not necessarily be limited by Figure 41 C in this respect.

In addition, audio coding apparatus 540C can be similar to audio coding apparatus 540B.However, extraction unit 542 can be from position Flow 517 rather than the V of rearranged sequenceT Q_DISTVector 539 or VT DISTVector 525E is (such as relative to the audio coding apparatus of Figure 40 510 Described situation) determine one or more described VT SMALLVector 521.As a result, extraction unit 542 can be by VT SMALLVector 521 passes It is delivered to mathematical units 546.

In addition, extraction unit 542 can determine the modified background spherical harmonic coefficient 515B' through audio coding from bit stream 517, it will These coefficients 515B' is transmitted to audio decoding unit 544, and audio decoding unit 544 can the encoded modified back of audio decoder Scape spherical harmonic coefficient 515B is to restore modified background spherical harmonic coefficient 537.Audio decoding unit 544 can be by these modified background balls Humorous coefficient 537 is transmitted to mathematical units 546.

Mathematical units 546 can be then by (and may be unordered) U through audio decoderDIST*SDISTVector 527' is multiplied by described one Or multiple VT SMALLVector 521 is to restore the different spherical harmonic coefficient of higher-order.Mathematical units 546 can be then humorous by the different ball of higher-order Coefficient be added with modified background spherical harmonic coefficient 537 with restore the multiple spherical harmonic coefficient 511 or its it is a certain export (it can be to return Because in the export of the Degree reduction executed at cell encoder 510E).

By this method, the technology can make audio decoding apparatus 540C can determine from background spherical harmonic coefficient recombinate to subtract In one or more vectors that the spherical harmonic coefficient of the amount of few position for needing to distribute to one or more vectors in bit stream decomposes at least One, wherein the spherical harmonic coefficient describes sound field, and wherein the background spherical harmonic coefficient describes one or more back of identical sound field Scape component.

In this respect, the various aspects of the technology can enable audio decoding apparatus 540C be configured in some cases To determine at least one of one or more vectors decomposed from the spherical harmonic coefficient recombinated with background spherical harmonic coefficient from bit stream, wherein The spherical harmonic coefficient describes sound field, and wherein the background spherical harmonic coefficient describes one or more background components of same sound field.

In these and other example, audio decoding apparatus 540C is configured to obtain having for spherical harmonic coefficient etc. from bit stream In NBGOrder first part.

In these and other example, audio decoding apparatus 540C is further configured to obtain spherical harmonic coefficient from bit stream With equal to NBGOrder the first part through audio coding, and first through audio coding of audio decoder spherical harmonic coefficient Divide to generate the first part of spherical harmonic coefficient.

In these and other example, in one or more described vectors it is described at least one include one or more VT SMALL Vector, one or more described VT SMALLVector is from by executing the singular value decomposition generation relative to the multiple spherical harmonic coefficient V matrix transposition determine.

In these and other example, in one or more described vectors it is described at least one include one or more VT SMALL Vector, one or more described VT SMALLVector is from by executing the singular value decomposition relative to the multiple spherical harmonic coefficient The transposition of the V matrix of generation determines, and audio decoding apparatus 540C is further configured to obtain from bit stream from U matrix and S One or more U derived from matrixDIST*SDISTVector, described two matrixes are all by executing relative to the multiple spherical harmonic coefficient Singular value decomposition and generate, and by UDIST*SDISTVector is multiplied by VT SMALLVector.

In these and other example, in one or more described vectors it is described at least one include one or more VT SMALL Vector, one or more described VT SMALLVector is from by executing the singular value decomposition relative to the multiple spherical harmonic coefficient The transposition of the V matrix of generation determines, and audio decoding apparatus 540C is further configured to obtain from bit stream from U matrix and S One or more U derived from matrixDIST*SDISTVector, described two matrixes are all by executing relative to the multiple spherical harmonic coefficient Singular value decomposition and generate, and by UDIST*SDISTVector is multiplied by VT SMALLVector to restore the different background spherical harmonic coefficient of higher-order, And the background spherical harmonic coefficient background spherical harmonic coefficient different with higher-order comprising the different background spherical harmonic coefficient of lower-order is added so that Small part restores the multiple spherical harmonic coefficient.

In these and other example, in one or more described vectors it is described at least one include one or more VT SMALL Vector, one or more described VT SMALLVector is from by executing the singular value decomposition relative to the multiple spherical harmonic coefficient The transposition of the V matrix of generation determines, and audio decoding apparatus 540C is further configured to obtain from bit stream from U matrix and S One or more U derived from matrixDIST*SDISTVector, described two matrixes are all by executing relative to the multiple spherical harmonic coefficient Singular value decomposition and generate, by UDIST*SDISTVector is multiplied by VT SMALLVector, will to restore the different background spherical harmonic coefficient of higher-order Background spherical harmonic coefficient background spherical harmonic coefficient different with higher-order comprising the different background spherical harmonic coefficient of lower-order is added at least portion Divide and restore the multiple spherical harmonic coefficient, and reproduces the multiple spherical harmonic coefficients restored.

Figure 41 D is the block diagram for illustrating another exemplary audio code device 540D.Audio decoding apparatus 540D can indicate energy Any device of enough decoding audio datas, for example, it is desktop PC, laptop computer, work station, tablet computer, dedicated Audio recording device, cellular phone (including so-called " smart phone "), personal media player apparatus, personal gaming device Or it is capable of the device of any other type of decoding audio data.

In the example of Figure 41 D, audio decoding apparatus 540D was executed and by being appointed in audio coding apparatus 510B to 510E The reciprocal audio decoding process of the audio encoding process that one executes, exception are to execute Degree reduction (as above relative to figure Described by the example of 40B to 40J), in some instances, the audio decoding process is used by audio coding apparatus 510B to 510J To facilitate the removal of external uncorrelated data.

Although being shown as single device, that is, the device 540D in the example of Figure 41 D, mentioned below is to be contained in dress Setting various assemblies or unit in 540D may be formed at isolated system outside device 540D.In other words, although in the present invention It is described as being executed by single device (that is, device 540D in the example of Figure 41 D), but the technology can be by including multiple devices System implement or in addition execute, wherein each of these devices can respectively contain various groups be described more particularly below One or more of part or unit.Therefore, the example that the technology should not necessarily be limited by Figure 41 D in this respect.

In addition, audio decoding apparatus 540D can be similar to audio decoding apparatus 540B, the difference is that audio decoding apparatus 540D is executed to be decompressed with the substantially reciprocal additional V that compresses executed by the V compression unit 552 described above in regard to Figure 40 I. In the example of Figure 41 D, extraction unit 542 includes V decompression unit 555, execute included in bit stream 517 (and substantially root It is specified according to the example shown in one of Figure 10 B and 10C) this V decompression through compression space component 539'.V decompression Contracting unit 555 can decompress V based on following equationT DISTVector 539:

In other words, V decompression unit 555 can dissect the nbits value from bit stream 517 first, and identify five Houghs The appropriate set of graceful code table when decoding the Huffman code of cid to use.It is specified suddenly based on prediction mode and in bit stream 517 Other order of element numbers of the element of the graceful decoding information of husband and possible space component relative to spatial component, V decompression unit 555 It can recognize the correct person in five huffman tables defined for the nbits value through dissecting.Use this huffman table, V decompression Unit 555 can decode cid value from Huffman code.V decompression unit 555 then can dissect sign bits and remaining block code, to residual Remaining block code is decoded to identify remnants.According to above equation, 555 decodable code V of V decompression unitT DISTOne in vector 539 Person.

Foregoing teachings can be summarized in following syntax table:

The decoded vector of table-

In aforementioned syntax table, tool there are four types of situation (situation 0-3) first switching sentence provide with regard to coefficient number and It says and determines VT DISTThe mode of vector length.First situation, situation 0, instruction are used for VT DISTAll coefficients of vector are designated.The Two situations, situation 1 indicate only VT DISTCorresponding to for vector be greater than those of the order of MinNumOfCoeffsForAmbHOA Number is designated, and can indicate to be referred to as (N aboveDIST+1)-(NBG+ 1) the case where.Third situation, situation 2 are similar to the second feelings Condition, but the coefficient of identification is further subtracted into NumOfAddAmbHoaChan, it indicates to correspond to over order N for specifiedBG Order additional channels (wherein " channel " refers to the particular factor corresponding to a certain order, sub- order combinations) variable.4th Situation, situation 3 indicate only VT DISTRemaining that after removing the coefficient that is identified by NumOfAddAmbHoaChan of vector A little coefficients are designated.

After switching sentence herein, if execute the decision of unified de-quantization by NbitsQ (or as indicated above, nbits) control System causes if NbitsQ is not equal to 5 using Hofmann decoding.Cid value mentioned above is equal to two of NbitsQ value most Low order.Prediction mode discussed herein above is expressed as PFlag in the above syntax table, and HT information bit is in the above syntax table In be expressed as CbFlag.The specified decoding of remaining grammer occurs as being how substantially similar to the mode of mode as described above.

By this method, technology of the invention can enable audio decoding apparatus 540D obtain the spatial component including sound field Bit stream through compressed version, the spatial component are produced and executing the synthesis based on vector relative to multiple spherical harmonic coefficients It is raw, and decompress spatial component through compressed version to obtain spatial component.

In addition, the technology aloow audio decoding apparatus 540D decompress sound field spatial component through compressed version This, the spatial component is generated and executing the synthesis based on vector relative to multiple spherical harmonic coefficients.

By this method, the various sides of the executable technology illustrated below in relation to following clause of audio coding apparatus 540D Face.

Clause 141541-1B.A kind of device comprising:

One or more processors, one or more described processors be configured to obtain include sound field spatial component through pressing The bit stream of contracting version, the spatial component be generated and executing synthesis based on vector relative to multiple spherical harmonic coefficients, and Decompress spatial component through compressed version to obtain spatial component.

Clause 141541-2B.The device of clause 141541-1B, wherein spatial component at least partly makes through compressed version It is shown in bit stream with the field of the specified prediction mode used when compression space component, and wherein one or more described processing Device is further configured to be at least partially based on the prediction mode decompression when decompression spatial component is through compressed version Spatial component through compressed version to obtain spatial component.

Clause 141541-3B.The device of any combination of clause 141541-1B and clause 141541-2B, wherein space is divided The huffman table information table for the huffman table of amount at least partly used using specified when compression space component through compressed version It is shown in bit stream, and wherein one or more described processors are further configured to when decompression spatial component is through compressed version When, be at least partially based on huffman table information decompression spatial component through compressed version.

Clause 141541-4B.The device of any combination of clause 141541-1B to clause 141541-3B, wherein space is divided The value for the quantization step or its variable of amount at least partly used using instruction expression when compression space component through compressed version Field be shown in bit stream, and wherein one or more described processors are further configured to the warp when decompression spatial component When compressed version, be at least partially based on described value decompression spatial component through compressed version.

Clause 141541-5B.The device of clause 141541-4B, wherein described value includes nbits value.

Clause 141541-6B.The device of any combination of clause 141541-4B and clause 141541-5B, wherein the position Stream includes multiple spatial components of sound field (comprising its spatial component through compressed version) through compressed version, and wherein described value Express the quantization step or its variable that use when compress the multiple spatial component, and wherein described in one or more processors into One step is configured to be at least partially based on described value when decompression spatial component is through compressed version and decompress the space point That measures is the multiple through compressed version.

Clause 141541-7B.The device of any combination of clause 141541-1B to clause 141541-6B, wherein space is divided The classification identification of the compression classification at least partly using Huffman code to indicate that identification spatial component corresponds to through compressed version of amount Symbol is to be shown in bit stream, and wherein one or more described processors are further configured to when decompression spatial component is through pressing When contracting version, be at least partially based on Huffman code decompression spatial component through compressed version.

Clause 141541-8B.The device of any combination of clause 141541-1B to clause 141541-7B, wherein space is divided Amount through compressed version at least partly using identification spatial component be on the occasion of or the sign bits of negative value be shown in bit stream, and Wherein one or more described processors are further configured to when decompression spatial component is through compressed version, at least partly base In sign bits decompression spatial component through compressed version.

Clause 141541-9B.Clause 141541-1B is to the device for any combination for arriving money 141541-8B, and wherein space is divided Amount at least partly indicates the residual value of spatial component using Huffman code to be shown in bit stream through compressed version, and wherein institute One or more processors are stated further to be configured to be at least partially based on Hough when decompression spatial component is through compressed version Graceful code decompression spatial component through compressed version.

Clause 141541-10B.The device of any combination of clause 141541-1B to clause 141541-10B, wherein being based on The synthesis of vector includes singular value decomposition.

In addition, audio decoding apparatus 540D can be configured to perform the various sides of the technology illustrated below in relation to following clause Face.

Clause 141541-1C.A kind of device, such as audio decoding apparatus 540D comprising: one or more processors, institute State one or more processors be configured to decompression sound field spatial component through compressed version, the spatial component is to pass through phase Synthesis based on vector is executed for multiple spherical harmonic coefficients and is generated.

Clause 141541-2C.The device of any combination of clause 141541-1C and clause 141541-2C, wherein described one Or multiple processors be further configured to when decompress spatial component through compressed version when, obtain identification spatial component when through The class identifier for the classification being categorized into when compression, obtain identification spatial component whether be positive value or negative value sign, obtain With spatial component through the associated residual value of compressed version, and based on class identifier, sign and residual value decompression Spatial component through compressed version.

Clause 141541-3C.The device of clause 141541-2C, wherein one or more described processors are through further through matching It sets when obtaining class identifier, to obtain the Huffman code for indicating class identifier, and the decoding Huffman code to obtain Class identifier.

Clause 141541-4C.The device of clause 141541-3C, wherein one or more described processors are further configured When decoding Huffman code, to be at least partially based on relative position of the spatial component in the vector for specifying multiple spatial components Identify the huffman table to decoding Huffman code.

Clause 141541-5C.The device of any combination of clause 141541-3C and clause 141541-4C, wherein described one Or multiple processors are further configured to when decoding Huffman code, and the use when compression space component is at least partially based on Prediction mode identify huffman table to decoding Huffman code.

Clause 141541-6C.The device of any combination of clause 141541-3C to clause 141541-5C, wherein described one Or multiple processors are further configured to when decoding Huffman code, are at least partially based on spatial component through compressed version Associated huffman table information identifies the huffman table to decoding Huffman code.

Clause 141541-7C.The device of clause 141541-3C, wherein one or more described processors are further configured With when decoding Huffman code, be at least partially based on relative position of the spatial component in the vector for specifying multiple spatial components, The prediction mode that is used when compression space component and know with spatial component through the associated huffman table information of compressed version Huffman table not to decoding Huffman code.

Clause 141541-8C.The device of clause 141541-2C, wherein one or more described processors are further configured With when obtaining residual value, decoding indicates the block code of residual value to obtain residual value.

Clause 141541-9C.The device of any combination of clause 141541-1C to clause 141541-8C, wherein based on to The synthesis of amount includes singular value decomposition.

In addition, audio decoding apparatus 540D can be configured to perform the various sides of the technology illustrated below in relation to following clause Face.

Clause 141541-1G.A kind of device, such as audio decoding apparatus 540D comprising: one or more processors, institute One or more processors are stated to be configured to based on multiple current spatial components through in compression space component through compressed version phase For the multiple order identification huffman codebook through the rest in compression space component to decompress the current sky Between component through being used when compressed version, the spatial component be by relative to multiple spherical harmonic coefficients execute the conjunction based on vector At and generate.

Clause 141541-2G.The device of clause 141541-1G, wherein one or more described processors are further configured To execute the step described in clause 141541-1D to clause 141541-10D and clause 141541-1E to clause 141541-9E Rapid any combination.

The Degree reduction unit 528A shown in the example of respective for more details Figure 40 B to the 40J of Figure 42 to 42C Block diagram.Figure 42 is the block diagram for illustrating Degree reduction unit 528, can indicate the Degree reduction unit 528A's of Figure 40 B to 40J One example.Degree reduction unit 528A can receive or be otherwise determined that targeted bit rates 535, and be based only upon this targeted bit rates 535 execute the Degree reduction relative to background spherical harmonic coefficient 531.In some instances, mesh can be used in Degree reduction unit 528A 535 access list of target rate or other data structures are to identify to those of remove order and/or son from background spherical harmonic coefficient 531 Order is to generate the background spherical harmonic coefficient 529 of reduction.

By this method, the technology can make audio coding apparatus (such as audio coding apparatus 510B to 410J) that can be based on Targeted bit rates 535 are executed relative to multiple spherical harmonic coefficients or its Degree reduction for decomposing (for example, background spherical harmonic coefficient 531), To generate the spherical harmonic coefficient 529 of reduction or the decomposition of its reduction, wherein the multiple spherical harmonic coefficient indicates sound field.

In each of various examples as described above, it should be appreciated that 540 executing method of audio decoding apparatus or Also comprise the device for executing each step that audio decoding apparatus 540 is configured to the method executed.In some cases, this A little devices may include one or more processors.In some cases, one or more described processors can indicate to be arrived by means of storage The application specific processor of the instruction configuration of non-transitory computer-readable storage media.In other words, in the set of encoding example The various aspects of technology in each can provide the non-transitory computer-readable storage media for being stored thereon with instruction, described Instruction causes one or more described processors to execute the method that audio decoding apparatus 540 has been configured to execute when being executed.

Figure 42 B is the block diagram for illustrating Degree reduction unit 528B, can indicate the Degree reduction unit of Figure 40 B to 40J An example of 528A.In the example of Figure 42 B, it is not based on only targeted bit rates 535 and executes Degree reduction, Degree reduction list First 528B can execute Degree reduction based on the content analysis of background spherical harmonic coefficient 531.Degree reduction unit 528B may include executing The content analysis unit 536A of this content analysis.

In some instances, content analysis unit 536A may include the content analysis for executing the form with reference to spatial analysis Spatial analysis unit 536A.Spatial analysis can be related to analysis background spherical harmonic coefficient 531 with the background component of identification description sound field The spatial information of shape or other spatial properties.Based on this spatial information, Degree reduction unit 528B be can recognize to from background ball Humorous coefficient 531 those of removes order and/or sub- order to generate the background spherical harmonic coefficient 529 of reduction.

In some instances, content analysis unit 536A may include the content analysis for executing the form for being known as diffusion analysis Diffusion analysis unit 536B.Diffusion analysis can be related to analysis background spherical harmonic coefficient 531 with the background component of identification description sound field Diffusible diffusion information.Information is spread based on this, Degree reduction unit 528B can recognize to remove from background spherical harmonic coefficient 531 Those of order and/or sub- order to generate the background spherical harmonic coefficient 529 of reduction.

Although being shown as comprising both spatial analysis unit 536A and diffusion analysis unit 36B, content analysis unit 536A may include only spatial analysis unit 536, only diffusion analysis unit 536B or spatial analysis unit 536A and diffusion analysis list Both first 536B.In some instances, other than one or both in spatial analysis and diffusion analysis, or substitution spatial analysis And one or both in diffusion analysis, the content analysis of other forms can be performed in content analysis unit 536A.Therefore, in this regard In technology described in the present invention should not be restricted.

By this method, the technology can make audio coding apparatus (such as audio coding apparatus 510B to 510J) that can be based on Describe sound field multiple spherical harmonic coefficients or its decompose content analysis execute relative to the multiple spherical harmonic coefficient or its decompose Degree reduction is to generate the spherical harmonic coefficient of reduction or the decomposition of its reduction.

In other words, the technology can make device (such as audio coding apparatus 510B to 510J) can be according to the following terms Configuration.

Clause 133146-1E.Any one of a kind of device, such as audio coding apparatus 510B to 510J comprising one Or multiple processors, one or more described processors are configured in multiple spherical harmonic coefficients based on description sound field or its decomposition Hold analysis execute relative to the multiple spherical harmonic coefficient or its decompose Degree reduction with generate reduction spherical harmonic coefficient or its contracting The decomposition subtracted.

Clause 133146-2E.The device of clause 133146-1E, wherein one or more described processors are further configured The different of description sound field is identified before executing Degree reduction, to execute the singular value decomposition relative to the multiple spherical harmonic coefficient One or more secondary vectors of the background component of one or more primary vectors and identification sound field of component, and it is wherein described one or more A processor be configured to execute relative to one or more described primary vectors, one or more described secondary vectors or described one or The Degree reduction of both multiple primary vectors and one or more secondary vectors.

Clause 133146-3E.The device of clause 133146-1E, wherein one or more described processors are further configured To execute relative to the multiple spherical harmonic coefficient or its content analysis decomposed.

Clause 133146-4E.The device of clause 133146-3E, wherein one or more described processors are configured to execute Relative to the multiple spherical harmonic coefficient or its spatial analysis decomposed.

Clause 133146-5E.The device of clause 133146-3E, wherein executing content analysis includes executing relative to described Multiple spherical harmonic coefficients or its diffusion analysis decomposed.

Clause 133146-6E.The device of clause 133146-3E, wherein one or more described processors are configured to execute Relative to the multiple spherical harmonic coefficient or its spatial analysis and diffusion analysis for decomposing.

Clause 133146-7E.The device of technical solution 1, wherein one or more described processors are configured to based on described Multiple spherical harmonic coefficients or its content analysis decomposed and targeted bit rates are executed relative to the multiple spherical harmonic coefficient or its decomposition Degree reduction with generate reduction spherical harmonic coefficient or its reduction decomposition.

Clause 133146-8E.The device of clause 133146-1E, wherein one or more described processors are further configured With the spherical harmonic coefficient of audio coding reduction or its decomposition.

Clause 133146-9E.The device of clause 133146-1E, wherein one or more described processors are further configured With the spherical harmonic coefficient of audio coding reduction or the decomposition of its reduction, and bit stream is generated comprising the spherical harmonic coefficient of reduction or its reduction Decomposition.

Clause 133146-10E.The device of clause 133146-1E, wherein one or more described processors are further configured One or more of the spherical substrate function corresponded to decomposition of the spherical harmonic coefficient for specifying in bit stream those to reduce or its reduction Order and/or one or more sub- orders, the bit stream include the spherical harmonic coefficient of the reduction or the decomposition of its reduction.

Clause 133146-11E.The device of clause 133146-1E, wherein through reduction spherical harmonic coefficient or its through reduction decompose tool There is the value than the multiple spherical harmonic coefficient or its decomposition of little.

Clause 133146-12E.The device of clause 133146-1E, wherein one or more described processors are further configured There is specified those of order and/or sub- order spherical harmonic coefficient in vector to remove the multiple spherical harmonic coefficient or its decomposition Vector with generate through reduction spherical harmonic coefficient or its through reduction decompose.

Clause 133146-13E.The device of clause 133146-1E, wherein one or more described processors are configured to institute State multiple spherical harmonic coefficients or its those of decompose in vector have specified those of order and/or sub- order spherical harmonic coefficient or to Amount pulverised with generate through reduction spherical harmonic coefficient or its through reduction decompose.

Figure 42 C is the block diagram for illustrating Degree reduction unit 528C, can indicate the Degree reduction unit of Figure 40 B to 40J An example of 528A.The Degree reduction unit 528C of Figure 42 B is substantially identical as Degree reduction unit 528B, but more than can pressing Relative to Figure 42 Degree reduction unit 528A describe mode receive or be otherwise determined that targeted bit rates 535, while also by with On relative to Figure 42 B Degree reduction unit 528B describe mode execute content analysis.Degree reduction unit 528C can be then The Degree reduction relative to background spherical harmonic coefficient 531 is executed based on this targeted bit rates 535 and content analysis.

By this method, the technology can be such that audio coding apparatus (such as audio coding apparatus 510B to 510J) is able to carry out Relative to the multiple spherical harmonic coefficient or its content analysis decomposed.When executing Degree reduction, audio coding apparatus 510B is arrived 510J can be executed based on targeted bit rates 535 and content analysis relative to the multiple spherical harmonic coefficient or its Degree reduction decomposed Reduced spherical harmonic coefficient with generation or it is decomposed through reduction.

It is assumed that removing one or more vectors, then audio coding apparatus 510B to 510J can be in bit stream by the number of vector It is appointed as control data.Audio coding apparatus 510B to 510J can refer to this fixed number of vectors to facilitate by audio solution in bit stream Code device is from bit stream extracted vector.

Figure 44 be illustrate according to the present invention described in the various aspects of technology executed by audio coding apparatus 410D To compensate the figure of the example operation of quantization error.In the example of Figure 44, the mathematical units 526 of audio coding apparatus 510D are opened up Dashed boxes are shown as to indicate that mathematical operation can be executed by the mathematical units 526 of audio decoding apparatus 510D.

As Figure 44 example in show, mathematical units 526 can be first by UDIST*SDISTVector 527 is multiplied by VT DISTVector 525E (is expressed as " H to generate different spherical harmonic coefficientDISTVector 630 ").Mathematical units 526 can be then by HDISTVector 630 divided by VT DIST(it is expressed as " V to the quantified version of vector 525E againT Q_DISTVector 525G ").Mathematical units 526 can pass through determination VT Q_DISTThe pseudoinverse of vector 525G and then by HDISTVector is multiplied by VT Q_DISTThe pseudoinverse of vector 525G executes this division, thus Export UDIST*SDIST(it can be abbreviated as " USDIST" or " USDISTVector ") the version through error compensation.In the example of Figure 44 In, USDISTThe version through error compensation be represented by US* DISTVector 527'.By this method, the technology can effectively at least Quantization error is projected to US by partDISTVector 527, to generate US* DISTVector 527'.

Mathematical units 526 can be then from UDIST*SDISTVector 527 subtracts US* DISTVector 527' is to determine USERRVector 634 (it, which can indicate to be attributed to, projects to UDIST*SDISTAt least part of the error of quantization in vector 527).Mathematical units 526 can Then by USERRVector 634 is multiplied by VT Q_DISTVector 525G is to determine HERRVector 636.Mathematically, HERRVector 636 can be equivalent In USDISTVector 527-US* DISTVector 527', result then with VT DISTVector 525E is multiplied.Mathematical units 526 can then by HERRVector 636 is by by UBGVector 525D is multiplied by SBGVector 525B and then multiplied by VT BGThe background ball that vector 525F is calculated is humorous Coefficient 531 (in the example of Figure 44, is expressed as HBGVector 531) it is added.Mathematical units 526 can be by HERRVector 636 and HBGTo Amount 531 is added, so that at least part of quantization error is effectively projected to HBGCompensated H is generated in vector 531BG Vector 531'.By this method, at least part of quantization error can be projected to H by the technologyBGIn vector 531.

If Figure 45 and 45B be illustrate according to the present invention described in technology various aspects the stem portion from two frames The figure of interpolation subframe.In the example of Figure 45, first frame 650 and the second frame 652 are shown.First frame 650 may include can be through decomposing For U [1], S [1] and V'[1] spherical harmonic coefficient (" SH [1] ") of matrix.Second frame 652 may include spherical harmonic coefficient (" SH [2] ").This A little SH [1] and SH [2] can recognize the different frame of SHC 511 described above.

In the example of Figure 45 B, the decomposition unit 518 of the audio coding apparatus 510H shown in the example of Figure 40 H can Each of frame 650 and 652 are divided into four corresponding subframe 651A to 651D and 653A to 653D.Decomposition unit 518 can connect The first subframe 651A (being expressed as " SH [1,1] ") of frame 650 is resolved into U [1,1], S [1,1] and V [1,1] matrix, thus V [1,1] matrix 519' is output to interpolation unit 550.Decomposition unit 518 can then (indicate the second subframe 653A of frame 652 For " SH [2,1] ") U [1,1], S [1,1] and V [1,1] matrix are resolved into, so that V [2,1] matrix 519' is output to interpolation list Member 550.Decomposition unit 518 can also be by the SH of the SH [1,1] of SHC 11, SH [1,2], SH [1,3] and SH [1,4] and SHC 511 [2,1], SH [2,2], SH [2,3] and SH [2,4] are output to interpolation unit 550.

Interpolation unit 550 can then execute the interpolation identified at the bottom for being diagram shown in the example of Figure 45 B. That is, interpolation unit 550 can be based on V'[1,1] and V'[2,1] interpolation V'[l, 2].Interpolation unit 550 can also be based on V'[1,1] and V'[2,1] interpolation V'[1,3].In addition, interpolation unit 550 can also be based on V'[1,1] and V'[2,1] interpolation V'[1,4].In these Inserting can be related to V'[1,1] and V'[2, the 1] projection into spatial domain, as shown in the example of Figure 46 to 46E, then for when Between interpolation, then project back in the humorous domain of ball.

Interpolation unit 550 can followed by by SH [1,2] multiplied by (V'[1,2])-1Export U [1,2] S [1,2], pass through by SH [1,3] is multiplied by (V'[1,3])-1U [1,3] S [1,3] is exported, and by SH [Isosorbide-5-Nitrae] multiplied by (V'[1,4])-1Export U [1,4] S [1,4].Interpolation unit 550 then can reform frame by the form through decomposing, to export V matrix 519, s-matrix 519B and U matrix 519C。

Figure 46 A to 46E be illustrate according to the present invention described in technology interpolation multiple spherical harmonic coefficients through decomposing The figure of the cross section of the projection of one or more vectors of version.Figure 46 A illustrates to have passed through SVD process from from first frame The cross section of the projection of one or more primary vectors for the first V matrix 19' that the SHC 511 of one subframe is decomposed.Figure 46 B illustrates Through by SVD process from the SHC 511 of the first subframe from the second frame the 2nd V matrix 519' decomposed one or more second The cross section of the projection of vector.

Figure 46 C illustrates to indicate the throwing of one or more interpolated vectors of the V matrix 519A of the second subframe from first frame The cross section of shadow, in accordance with the techniques described in this disclosure from V matrix 519' interpolation, the V matrix 519' is these vectors From the first subframe of the first frame of SHC 511 (that is, V matrix 519''s shown in the example of Figure 46 in this example is one or more A vector) and SHC 511 the second frame the first subframe (that is, V matrix 519' shown in the example of Figure 46 B in this example One or more vectors) decompose.

Figure 46 D illustrates to indicate the throwing of one or more interpolated vectors of the V matrix 519A of the third subframe from first frame The cross section of shadow, in accordance with the techniques described in this disclosure from V matrix 519' interpolation, the V matrix 519' is these vectors From the first subframe of the first frame of SHC 511 (that is, V matrix 519''s shown in the example of Figure 46 in this example is one or more A vector) and SHC 511 the second frame the first subframe (that is, V matrix 519' shown in the example of Figure 46 B in this example One or more vectors) decompose.

Figure 46 E illustrates to indicate the throwing of one or more interpolated vectors of the V matrix 519A of the 4th subframe from first frame The cross section of shadow, in accordance with the techniques described in this disclosure from V matrix 519' interpolation, the V matrix 519' is these vectors From the first subframe of the first frame of SHC 511 (that is, V matrix 519''s shown in the example of Figure 46 in this example is one or more A vector) and SHC 511 the second frame the first subframe (that is, V matrix 519' shown in the example of Figure 46 B in this example One or more vectors) decompose.

Figure 47 is the extraction that audio decoding apparatus 540A to 540D shown in the example of Figure 41 to 41D is described in more detail The block diagram of unit 542.In some instances, extraction unit 542 can indicate to can be referred to the front end of " integrated decoder ", can hold Two or more decoding schemes of row are (wherein by executing the two or more than two schemes, it is believed that decoder " integrated " Described two or more than two schemes).As shown in the example of Figure 44, extraction unit 542 include multiplexer 620 and Extract subelement 622A and 622B (" extracting subelement 622 ").Multiplexer 620 is based on associated encoded frame formula SHC The encoded frame formula SHC matrix 547 of corresponding instruction identification that matrix 547 still is generated from record to 547N from Composite tone object to It is sent in 547N and extracts those of subelement 622A and extraction subelement 622B matrix.It extracts each in subelement 622A Different decodings (it can be referred to " decompression ") scheme can be performed in person, and in some instances, the scheme is adapted to from synthesis The SHC that audio object the generates or SHC generated from record.Extracting each of subelement 622A can be performed these decompression sides Corresponding one in case, to generate the frame of the SHC 547 to be output to SHC 547.

For example, decompression scheme can be performed to use following equation from main signal (PS) weight in extraction unit 622A Build SA:

HOA=DirV × PS,

Wherein DirV is direction vector (indicating various directions and width), can be transmitted via wing passage.In this example, The executable decompression scheme for rebuilding HOA matrix from PS using following equation of extraction unit 622B:

Wherein Ynm be spheric harmonic function, and θ andInformation can be sent via wing passage.

In in this respect, the technology enables extraction unit 538 based on the spherical harmonic coefficient for indicating sound field through compressed version One of originally whether select multiple decompression schemes from the instruction that Composite tone object generates, and use the multiple decompression Selected person in contracting scheme decompresses the described through compressed version of the spherical harmonic coefficient.In some instances, described device includes Integrated decoder.

Figure 48 is the sound that audio decoding apparatus 540A to 540D shown in the example of Figure 41 A to 41D is described in more detail The block diagram of frequency reproduction unit 48.Figure 48 explanation is from restored spherical harmonic coefficient 547 to simultaneous with decoder part loudspeaker geometrical arrangements The conversion of the multichannel audio data 549A of appearance.For some local loudspeaker geometrical arrangements, (it can be referred to again in decoder The loudspeaker geometrical arrangements at place), it is ensured that reversible some transformation can lead to less desirable audio image quality.Also It is to say, when compared with the audio captured, sound reproduction can be not always to lead to being properly positioned for sound.In order to this not Big desirable picture quality is corrected, and can further enhance technology to introduce the concept that can be referred to " virtual speaker ".

The above frame can be modified by comprising the translation in the form of a certain, for example, vector basal amplitude translation (VBAP), based on away from From amplitude translation or other forms translation, rather than need that one or more loudspeakers are repositioned or are positioned at have by example In the specific or defined area in the space of the specified special angle tolerance of such as above-mentioned ITU-R BS.775-1 standard.For The purpose of explanation and focus on VBAP, VBAP can actually introduce the thing for being characterized by " virtual speaker ".VBAP can The feeding of one or more loudspeakers is modified to so that these one or more loudspeakers actually export seems from virtual loudspeaking The sound of device, the virtual speaker be in the position for being different from supporting one or more loudspeakers of the virtual speaker and/ Or the position and one or more of angle place of at least one of angle.

To be illustrated, for determining that the following equation of loudspeaker feeding can be as follows according to SHC:

In equation above, VBAP matrix have M row multiply N column size, wherein M indicate loudspeaker number (and It will be equal to five) in above equation, and N indicates the number of virtual speaker.It can be according to from the defined position of listener to raising The vector of each of the position of sound device and from the defined position of listener to each of the position of virtual speaker Vector calculate VBAP matrix.D matrix in above equation can have N row to multiply (order+1)2The size of column, wherein order It can be referred to the order of SH function.D matrix can indicate following equation

G matrix (or in view of there is only single column, vectors) can indicate to be arranged in raising in decoder local geometric arrangement The gain of the speaker feeds of sound device.In equation, g matrix has size M.A matrix (or consider there is only single column, Vector) it can indicate SHC 520, and there is size (order+1) (order+1), (order+1) can also be expressed as2

In fact, VBAP matrix is MxN matrix, the adjustment that can be referred to " gain adjustment " is provided, the adjustment will raise The position of sound device and the position of virtual speaker are taken into account.Introducing translation by this method can lead to the preferable weight of multi-channel audio Existing, this leads to the good quality image when being regenerated by local loudspeaker geometrical arrangements.In addition, by the way that VBAP is incorporated into this side In formula, technology can overcome the bad loudspeaker geometrical arrangements not being aligned with geometrical condition specified in various standards.

In fact, the particular geometric arrangement for being used for loudspeaker can be inverted and be transformed back to SHC using the equation Or the multichannel feeding of configuration, the particular geometric arrangement or configuration can be referred to as decoder local geometric again in the present invention Arrangement.That is, equation can be inverted to solve g matrix.Equation through inverting can be as follows:

G matrix can indicate the loudspeaking of each of five loudspeakers being used in 5.1 speaker configurations in this example Device gain.Virtual loudspeaker positions used in this configuration can correspond to defined in 5.1 multi-channel format specifications or standard Position.It any number of known audio location technology that can be used can support each of these virtual speakers to determine The position of loudspeaker, many technologies in the technology, which are related to playing, has the tone of specific frequency with each loudspeaker phase of determination For head-end unit (such as audio/video receiver (A/V receiver), TV, game system, digital video disk system or Other types of head-end system) position.Alternatively, the user of head-end unit can manually specify the position of each of loudspeaker It sets.Under any circumstance, in the case where giving these known locations and possible angle, it is assumed that virtual loudspeakers pass through VBAP Desired configuration, the gain of head-end unit can be solved.

In in this respect, device or equipment can execute the translation of vector basal amplitude or other shapes to the multiple virtual channel The translation of formula, with generate drive decoder local geometric arrange in loudspeaker transmitting be rendered as originating from being configured to different offices Multiple channels of the sound of the virtual speaker of portion's geometrical arrangements.Therefore the technology aloows audio decoding apparatus 40 right The multiple spherical harmonic coefficient (such as restored spherical harmonic coefficient 47) executes transformation to generate multiple channels.In the multiple channel Each can not same district corresponding with space it is associated.In addition, each of the multiple channel may include multiple virtual logical Road, wherein the multiple virtual channel can not same district corresponding with space it is associated.Device therefore virtual channel can be executed to Amount basal amplitude is translated to generate the multiple channel of multichannel audio data 49.

Figure 49 A to 49E (ii) is to illustrate that the respective audio of the various aspects of implementable technology described in the present invention is translated The figure of code system 560A to 560C, 567D, 569D, 571E and 573E.As shown in the example of Figure 49 A, audio-frequency decoding system 560A may include audio coding apparatus 562 and audio decoding apparatus 564.Audio coding apparatus 562 can be similar to respectively in Fig. 4 and Any one of audio coding apparatus 20 and 510A to 510D shown in the example of 40A to 40D.Audio decoding apparatus 564 can Audio decoding apparatus 24 and 40 shown in example similar to Fig. 5 and 41.

As described above, higher-order ambiophony (HOA) is to describe owning for the sound field based on spatial Fourier transform The mode of directional information.In some instances, ambiophony order N is higher, and space spatial resolution is higher, and ball is humorous (SH) The number (N+1) of coefficient2It is bigger.Therefore, in some instances, ambiophony order N is higher, will lead to be used for transmission and The bandwidth requirement of packing coefficient is bigger.Since compared with such as 5.1 or 7.1 surround sound audio data, the bandwidth requirement phase of HOA Work as height, so can desired bandwidth reduction for many applications.

Technology described according to the present invention, audio-frequency decoding system 560A is executable to be based on inciting somebody to action in spatial sound sound field scape The method of different (prospect) and non-different (background or environment) element sepatation.This, which is separated, allows audio-frequency decoding system 560A each other It is independently processed from prospect and background element.In this example, audio-frequency decoding system 560A Utilization prospects element can attract more notes The property of meaning (by hearer), and positioning can be easier to compared with background element (again by hearer).Therefore, audio-frequency decoding system 560A can more effectively storage or transmission HOA content.

In some instances, audio-frequency decoding system 560A can realize this point by using singular value decomposition (SVD) process From.The frame of HOA coefficient can be separated into 3 matrixes (U, S, V) by SVD process.Matrix U contains left singular vector, and V matrix contains Right singular vector.Diagonal matrix S contains non-negative categorized singular value in its diagonal line.HOA coefficient it is usual it is good (or Under some cases, perfection assumes to indicate the infinite precision in HOA coefficient) it rebuilds and will be provided by U*S*V'.Only by maximum odd with D Different value reconstruction subspace: U (:, 1:D) and * S (1:D :) * V', audio-frequency decoding system 560A can extract most of prominent from this HOA frame Spatial information, i.e. foreground sounds element (and can be some strong early stage indoor reflections).Rest part U (:, D+1:end) * S (D+1: End :) * V' can be from the content reconstruction background element and reverberation.

Audio-frequency decoding system 560A, which can be determined by analysis by the slope of a curve of the diagonal line value generation of successively decreasing of S, makes two The value D of subspace separation: big singular value indicates that foreground sounds, low singular value indicate background value.Audio-frequency decoding system 560A can make With the single order and second dervative of singular value curve.Number D can be also limited between one and five by audio-frequency decoding system 560A.Or Number D can be predefined as such as value four by person, audio-frequency decoding system 560A.Under any circumstance, once estimating number D, sound Frequency decoding system 560A extracts prospect and background subspace from matrix U and S.

Audio-frequency decoding system 560A then can rebuild background scene via U (:, D+1:end) * S (D+1:end :) * V' HOA coefficient, so as to cause (N+1) of HOA coefficient2Channel.Since it is known background element is not same in some instances prominent And not can equally be positioned relative to foreground elements, so the order in the channel HOA can be truncated in audio-frequency decoding system 560A.This Outside, audio-frequency decoding system 560A can be with damaging or lossless audio codec (such as AAC) compresses these channels, or optionally uses More positive audio codec is compressed compared with the codec to compress prominent foreground elements.In some cases Under, in order to save bandwidth, audio-frequency decoding system 560A can differently transmit foreground elements.That is, audio-frequency decoding system can be with having Damage or lossless audio codec (such as AAC) transmit left singular vector U (:, 1:D) after being compressed, and together with reconstruction square Battle array R=S (1:D :) * V' transmits these through compressing left singular value together.R can indicate D × (N+1)2Matrix, the matrix can be across Frame and it is different.

At the receiver-side of audio-frequency decoding system 560, audio-frequency decoding system can make the two matrix multiples to rebuild (N+ 1)2The frame in a channel HOA.Once background and the channel prospect HOA are summed together, audio-frequency decoding system 560A can be used any Appropriate ambiophony reconstructor is reproduced to the setting of any loudspeaker.Because the technology provides foreground elements (direct or phase Allophone sound) with the separation of element, so the people of hearing impairment can control prospect to the mixing of background element to increase intelligibility. Also, other audio frequency effects can also be applicable, such as only to the dynamic compressor of foreground elements.

Figure 49 B is the block diagram that audio coding system 560B is described in more detail.As shown in the example of Figure 49 B, audio is translated Code system 560B may include audio coding apparatus 566 and audio decoding apparatus 568.Audio coding apparatus 566 can be similar to Fig. 4 and Audio coding apparatus 24 and 510E shown in the example of 40E.Audio decoding apparatus 568 can be similar in the example of Fig. 5 and 41B Shown in audio decoding apparatus 24 and 540B.

Technology described according to the present invention, when the purpose reduced for bandwidth uses based on frame HoA signal When SVD (or correlation technique, such as KLT&PCA) is decomposed, audio coding apparatus 66 can quantify preceding several vectors of U matrix (multiplied by S The correspondence singular value of matrix) and VTThe correspondence vector of vector.This is by ' prospect ' component including sound field.The technology can make sound Frequency code device 566 is able to use ' black box ' audio coding engine to UDIST*SDISTVector is decoded.V vector can be through scalar quantity Change or vector quantization.In addition, some or all of the residual vector in U matrix can be unusual multiplied by the correspondence of s-matrix and V matrix Value and also use ' black box ' audio coding engine decoded.These are by ' background ' component including sound field.

Because most loud acoustic components are resolved into ' prospect component ', audio coding apparatus 566 can be at use ' black box ' The ambiophony order of ' background ' component is reduced before audio coding engine, because (it is assumed that) background and without containing important Content can be positioned.Depending on the ambiophony order of prospect component, audio coding unit 566 can transmit corresponding V vector, corresponding V Vector can be quite big.For example, simple 16 scalar quantizations of V vector will generate each prospect component and about be used for the 4th rank 20kbps expense (25 coefficients) and 40kbps (49 coefficients) for the 6th rank.Technology described in the present invention can mention For the method for this expense to reduce V vector.

To be illustrated, it is assumed that the ambiophony order of foreground elements is NDIST, and the ambiophony order of background element For NBG, as described above.Because audio coding apparatus 566 can reduce the ambiophony rank of background element as described above Number, so NBGIt is smaller than NDIST.It needs to be transmitted to rebuild the length of the prospect V vector of foreground elements at receiver-side and have Each foreground elements (NDIST+1)2Length, and the first ((NDIST+1)2)-((NBG+1)2) coefficient can be used to rebuild be up to rank Number NBGProspect or distinct components.Using technology described in the present invention, it is at most order that audio coding apparatus 566, which can be rebuild, NBGProspect and by gained (NBG+1)2A channel merges with background channel, so as to cause being at most order NBGFull sound field. V vector can then be reduced to have by audio coding apparatus 566 is higher than (NBG+1)2Index those of to be used for transmission coefficient (wherein these vectors can be referred to " VT SMALL").At receiver-side, audio decoding unit 568 can by by foreground elements with VT SMALLMultiplication of vectors is used to be greater than N to rebuildBGAmbiophony order prospect voice-grade channel.

Figure 49 C is the block diagram that audio coding system 560C is described in more detail.As shown in the example of Figure 49 C, audio is translated Code system 560B may include audio coding apparatus 567 and audio decoding apparatus 569.Audio coding apparatus 567 can be similar to Fig. 4 and Audio coding apparatus 20 and 510F shown in the example of 40F.Audio decoding apparatus 569 can be similar in the example of Fig. 5 and 41B Shown in audio decoding apparatus 24 and 540B.

Technology described according to the present invention, when the purpose reduced for bandwidth uses based on frame HoA signal When SVD (or correlation technique, such as KLT&PCA) is decomposed, audio coding apparatus 567 can quantify U matrix preceding several vectors (multiplied by The correspondence singular value of s-matrix) and VTThe correspondence vector of vector.This is by ' prospect ' component including sound field.The technology can make Audio coding apparatus 567 is able to use ' black box ' audio coding engine to UDIST*SDISTVector is decoded.V vector can be through scalar Quantization or vector quantization.In addition, some or all of the residual vector in U matrix can be multiplied by s-matrix and the corresponding surprise of V matrix Different value and also use ' black box ' audio coding engine decoded.These are by ' background ' component including sound field.

Because most loud acoustic components are resolved into ' prospect component ', audio coding apparatus 567 can be at use ' black box ' The ambiophony order of ' background ' component is reduced before audio coding engine, because (it is assumed that) background and without containing important Content can be positioned.Audio coding apparatus 567 can make the gross energy for retaining the sound field according to technology described herein Mode reduces order.Depending on the ambiophony order of prospect component, audio coding unit 567 can transmit corresponding V vector, corresponding V vector can be quite big.For example, simple 16 scalar quantizations of V vector will generate each prospect component and about be used for the 4th rank 20kbps expense (25 coefficients) and 40kbps (49 coefficients) for the 6th rank.Technology described in the present invention can mention For the method for this expense to reduce V vector.

To be illustrated, it is assumed that the ambiophony order of foreground elements and background element is N.Audio coding apparatus 567 can The ambiophony order of the background element of V vector is set to be reduced to η from N, so that η < N.Audio coding apparatus 67 is further applied Compensation is to increase the value of the background element of V vector, to retain the gross energy by the SHC sound field described.Above in relation to Figure 40 F Case technology of the description for application compensation.At receiver-side, audio decoding unit 569 can be rebuild for ambiophony order Background audio channel.

Figure 49 D (i) and 49D (ii) illustrate audio coding apparatus 567D and audio decoding apparatus 569D respectively.Audio coding Device 567D and audio decoding apparatus 569D can be configured to perform according to aspects of the present invention one or more be based on direction Property diversity determine.Higher-order ambiophony (HOA) is all directions letter for describing the sound field based on spatial Fourier transform The method of breath.In general, ambiophony order N is higher, spatial resolution is higher, and number (N+1) ^2 of humorous (SH) coefficient of ball is got over Greatly, bigger with bandwidth required for storing data for emitting.Because the bandwidth requirement of HOA is quite high, for many For, desired bandwidth is reduced.

Previous description has been described how SVD (singular value decomposition) or correlated process can be used for space audio compression.Herein Described in technology the innovatory algorithm for selecting prominent element (also referred to as foreground elements) is presented.By HOA audio frame base After SVD resolves into its U, S and V matrix, the technology makes the selection of K prominent element exclusively based on the preceding K of U matrix A channel [U (:, 1:K) * S (1:K, 1:K)].This causes selection with the audio element of highest energy.However, not guaranteeing that A little elements are also directionality.Therefore, the technology is to be directed to the sound element for being found to have high-energy and being also directionality.This It is potentially realized and being weighted to V matrix and s-matrix.Then, for every a line of this gained matrix, to higher rope Draw element (it is associated with higher-order HOA coefficient) it is squared and summation, so as to cause every a line one value [relative to figure SumVS in the pseudo-code of 40H description].According to the workflow indicated in pseudo-code, the higher-order started at the 5th index is considered Ambiophony coefficient.These values are classified according to its size, and classified index is correspondingly to rearrange original U, S and V square Battle array.Then it can be applicable to the relatively early described compression algorithm based on SVD in the present invention and modified without further.

Figure 49 E (i) and 49E (ii) is the block diagram for illustrating audio coding apparatus 571E and audio decoding apparatus 573E respectively. Described in the executable example above in relation to Figure 49 to 49D (ii) of audio coding apparatus 571E and audio decoding apparatus 573E The various aspects of technology, in addition to the executable power spectral density matrix (PDS) relative to HOA coefficient of audio coding apparatus 571E Singular value decomposition is to generate S2Except matrix and V matrix.S2Matrix can indicate squared s-matrix, therefore S2Matrix can undergo flat Root is operated to obtain s-matrix.In some cases, audio coding apparatus 571E can execute quantization relative to V matrix to obtain Quantified V matrix (it is represented by V' matrix).

Audio coding apparatus 571E can be by first obtaining s-matrix with quantified V' matrix multiple to generate SV' matrix Obtain U matrix.Next audio coding apparatus 571E can get the pseudoinverse of SV' matrix and then by HOA coefficient multiplied by SV' matrix Pseudoinverse is to obtain U matrix.Pass through the power spectral density relative to HOA coefficient rather than coefficient itself execution SVD, audio coding apparatus 571E can potentially reduce the computation complexity for executing SVD in terms of one or more of processor circulation and memory space, It is realized simultaneously as SVD directly applies to the identical source audio code efficiency of HOA coefficient.

Audio decoding apparatus 573E can be similar to those described above audio decoding apparatus, in addition to audio decoding apparatus 573 can be from the decomposition for the HOA coefficient realized by the power spectral density that SVD is applied to HOA coefficient rather than directly from HOA system Number is to rebuild except HOA coefficient.

Figure 50 A and 50B be respective explanation according to the present invention described in technology potentially reduce the order of background content One of two distinct methods block diagram.As shown in the example of Figure 50, first method be can be used relative to UBG*SBG*VT Order is reduced to η from N by the Degree reduction of vector, and wherein η is less than (<) N.That is, order shown in the example of Figure 40 B to 40J Reducing unit 528A can be performed Degree reduction to be truncated, or in other ways by UBG*SBG*VTThe order N of vector is reduced to η, wherein η is less than (<) N.

As an alternative, Degree reduction unit 528A can be as shown in the example of Figure 50 B relative to VTExecute this section It is disconnected to eliminate to go as (η+1)2, do not illustrate in the example of Figure 40 B for purpose of explanation and.In other words, order contracts Subtracting unit 528A can be removed VTOne or more orders of matrix are effectively to generate VBGMatrix.This VBGThe size of matrix is (η+1)2x(N+1)2- D, wherein this VBGMatrix is then as generation UBG*SBG*VTV is replaced when vectorTMatrix uses, to efficiently perform Truncation is to generate size M × (η+l)2UBG*SBG*VTVector.

Figure 51 is to illustrate the various aspects of implementable technology described in the present invention to compress the audio of spherical harmonic coefficient 701 The block diagram of the example of the distinct components compressed path of code device 700A.In the example of Figure 51, distinct components compressed path can Refer to the processing path for compressing the audio coding apparatus 700A by the distinct components of the sound field indicated of SHC 701.It can be referred to back Another path of scape component compressed path can indicate the processing road of the audio coding apparatus 700A of the background component of compression SHC 701 Diameter.

Although for the purpose of ease of explanation and it is not shown, background component compressed path can directly with respect to SHC 701 and It is not that the decomposition of SHC 701 is operated.It is similarly to above in relation to situation described in Figure 49 to 49C, in addition to background point Amount processing path can operate (audio coding apparatus 20 as shown in the example above in relation to Fig. 4 directly with respect to SHC 701 It is described), applied mental encoded acoustic device compresses these background components, rather than from UBG、SBGAnd VBGMatrix recombinates background point Amount and the psychologic acoustics coding (for example, using AAC encoder) for then executing a certain form that these recombinate background components.Pass through The psychologic acoustics executed directly with respect to SHC 701 encodes, and encodes phase relative to the psychologic acoustics of recombination background component with executing Compare, discontinuity can be reduced, while also reducing computation complexity (operating aspect required for compressed background component).Although In terms of being related to different and background, but in the present invention, term " significant " replacement " different " can be used, and term " environment " can be used Instead of " background ".

In any case, spherical harmonic coefficient 701 (" SHC 701 ") may include with M × (N+1)2Size coefficient square Battle array, wherein M indicates the number (and in some instances, 1024) of the sample in audio frame, and N indicates substrate corresponding to coefficient The top step number of function.As indicated above, for 1024 × 25 coefficients in total, N is typically set at four (4).Corresponding to spy Each of the SHC 701 for determining the sub- order combinations of order can be referred to channel.For example, correspond to the sub- rank substrate of single order zero All M sample coefficients of function can indicate channel, and the coefficient for corresponding to the sub- rank basis function of zeroth order zero can indicate another logical Road etc..SHC 701 is also known as higher-order ambiophony (HOA) content 701 or SH signal 701 in the present invention.

As shown in the example of Figure 51, audio coding apparatus 700A includes analytical unit 702, the synthesis unit based on vector 704, (" compression is single for vector reduction unit 706, psychologic acoustics coding unit 708, coefficient reduction unit 710 and compression unit 712 Member 712 ").Analytical unit 702 can indicate to be configured to execute the analysis relative to SHC 701 to identify sound field (D) 703 Distinct components and background component (BGTOT) 705 sum unit.Compared with audio coding apparatus as described above, audio Code device 700A does not execute this determination of the decomposition relative to SHC 701, but executes directly with respect to SHC 701 really It is fixed.

Synthesis unit 704 based on vector indicate to be configured to execute a certain form relative to SHC 701 based on to The synthesis (such as the synthesis of SVD, KLT, PCA or any other based on vector) of amount has M × (N to generate in the example of SVD +1)2Size [US] matrix 707 and have (N+1)2×(N+1)2Size [V] matrix 709 unit.[US] matrix 707 can indicate the matrix by generating by the matrix multiple that SVD is applied to [U] matrix and [S] matrix that SHC 701 is generated.

Vector reduction unit 706 can indicate to be configured to the number of vectors of reduction [US] matrix 707 and [V] matrix 709 with So that [US } matrix 707 and [V] matrix 709 each of the residual vector phase exclusive or significant component of identification sound field list Member.Vector, which reduces unit 706, can be performed this reduction of the number based on distinct components D 703.The number of distinct components D 703 can Number array is actually indicated, wherein the different vector of difference of each number recognition matrix 707 and 709.Vector reduces unit 706 Reduction [US] matrix 711 and size (N+1) of exportable size M × D2Reduction [V] matrix 713 of x D.

Although not shown for the purpose for being easy explanation, the interpolation of [V] matrix 709, which can be similar to, to be described in more detail above Mode mode [V] matrix 709 reduction before occur.In addition, although not shown for the purpose for being easy explanation, more than The mode that text is more fully described reorder reduction [US] matrix 711 and/or reduction [V] matrix 712.Therefore, the technology is not Should be restricted in these and other aspects (such as it is in error pro or example described above but in Figure 51 and (not shown) Any other aspect of aforementioned techniques).

Psychologic acoustics coding unit 708 indicate be configured to execute relative to [US] matrix 711 psychologic acoustics coding with Generate the unit of bit stream 715.Coefficient reduction unit 710 can indicate the number for being configured to the channel that reduction is reduced [V] matrix 713 Purpose unit.In other words, coefficient reduction unit 710 can indicate to be configured to eliminate have the different V of few directional information to Measure the unit of those of (its formation is reduced [V] matrix 713) coefficient.As described above, in some instances, correspond to one Those of the different V vector of rank and zeroth order basis function coefficient (is indicated hereinabove as NBG) provide few directional information and therefore can be from Different V vector removes (by the reduction for being referred to as " Degree reduction " above).In this example, it is possible to provide larger flexibility is not with Only from set [(NBG+1)2+ 1, (N+1)2] identify corresponding to NBGThis these coefficient and also (it can be by the identification additional channel HOA Variable TotalOfAddAmbHOAChan is indicated).Analytical unit 702 can analyze SHC 701 to determine BGTOT, not only can recognize (NBG+1)2And identification TotalOfAddAmbHOAChan.Coefficient reduction unit 710 can then be moved from [V] matrix 713 is reduced Except corresponding to (NBG+1)2And those of TotalOfAddAmbHOAChan coefficient, to generate size as ((N+1)2-(BGTOT)×D Small [V] matrix 717.

Scalar quantization noted above and/or huffman coding can be performed then to compress small [V] square in compression unit 712 Battle array 717, output are used as side channel information 719 (" side channel information 719 ") through compressing small [V] matrix 717.Compression unit 712 modes that can be similar to mode shown in the example of Figure 10 to 10O (ii) export side channel information 719.Some In example, side channel information 719 can be incorporated into bit stream 715 by the bitstream producing unit for being similar to that described above unit In.In addition, as noted above, audio coding apparatus 700A may include the back for generating another bit stream when being referred to as bit stream 715 Scape component processing path, wherein being similar to that described above the bitstream producing unit of unit can produce similar to described above Bit stream 17 bit stream, the bit stream includes bit stream 715 and the bit stream that is exported by background component processing path.

In accordance with the techniques described in this disclosure, the first non-zero that analytical unit 702 can be configured to determine the coefficient of vector Set, i.e., reduced the vector of [V] matrix 713, to the distinct components for indicating sound field in this example.In some examples In, analytical unit 702 can determine that side channel will be included in by forming all coefficients for each vector for being reduced [V] matrix 713 In information 719.Therefore analytical unit 702 can set BGTOTEqual to zero.

Audio coding apparatus 700A therefore can be effectively to be retouched with above in relation to the table for being expressed as " decoded vector " The reciprocal mode of the mode stated works.In addition, audio coding apparatus 700A can access unit header (its may include one or Multiple frames) in specify the syntactic element of any one in the multiple configuration mode of selection.Although depicted as in every access unit On the basis of specified, but analytical unit 702 can be in every frame basis or the basic (example of any other periodic basis or aperiodicity As entire bit stream it is primary) on specify this syntactic element.Under any circumstance, this syntactic element may include that instruction has selected four to match Setting which one in mode indicates the direction of this distinct components for the specified nonzero coefficient set for being reduced [V] matrix 713 Two positions of aspect.Institute's syntax elements are represented by " codedVVecLength ".By this method, audio coding apparatus 700A Any it can be indicated or be specified in other ways with signal in bit stream to determine using the one in four configuration modes to refer in bit stream Small [V] matrix 717.Although being described by relative to four configuration modes, the technology should not necessarily be limited by four configuration modes, and It should be limited to any number of configuration mode, include single configuration mode or multiple configuration modes.

Therefore, the various aspects of the technology can enable audio coding apparatus 700A be configured to be grasped according to following clause Make.

Clause 133149-1F.A kind of device comprising: one or more processors, one or more described processors are configured To select one of multiple configuration modes of non-null set of coefficient to specify vector, the vector is from description sound field And indicate that multiple spherical harmonic coefficients of the distinct components of sound field decompose, and institute is specified based on the selected person in the multiple configuration mode State the non-null set of the coefficient of vector.

Clause 133149-2F.The device of clause 133149-1F, wherein the one in the multiple configuration mode refers to The non-null set for showing the coefficient includes all coefficients.

Clause 133149-3F.The device of clause 133149-1F, wherein the one in the multiple configuration mode refers to The non-null set for showing coefficient includes corresponding in the coefficient to be greater than one or more of the multiple spherical harmonic coefficient institute Those of the order of order of corresponding basis function coefficient.

Clause 133149-4F.The device of clause 133149-1F, wherein the one in the multiple configuration mode refers to The non-null set for showing coefficient includes corresponding in the coefficient to be greater than one or more of the multiple spherical harmonic coefficient institute Those of the order of order of corresponding basis function coefficient, and exclude humorous corresponding to the multiple ball is greater than in the coefficient At least one coefficient of the order of the order of the corresponding basis function of the one or more in coefficient.

Clause 133149-5F.The device of clause 133149-1F, wherein the one in the multiple configuration mode refers to The non-null set for showing coefficient includes all coefficients, except at least one of described coefficient.

Clause 133149-6F.The device of clause 133149-1F, wherein one or more described processors are further configured To refer to the selected one in fixed the multiple configuration mode in bit stream.

Clause 133149-1G.A kind of device comprising: one or more processors, one or more described processors are configured It, will be according to the non-zero of the coefficient of the one extracted vector in multiple configuration modes with one of multiple configuration modes of determination Set, the vector are decomposed from multiple spherical harmonic coefficients of the distinct components of description sound field and expression sound field, and based on described The winner in multiple configuration modes specifies the non-null set of the coefficient of the vector.

Clause 133149-2G.The device of clause 133149-1G, wherein the one in the multiple configuration mode refers to The non-null set for showing the coefficient includes all coefficients.

Clause 133149-3G.The device of clause 133149-1G, wherein the one in the multiple configuration mode refers to The non-null set for showing coefficient includes corresponding in the coefficient to be greater than one or more of the multiple spherical harmonic coefficient institute Those of the order of order of corresponding basis function coefficient.

Clause 133149-4G.The device of clause 133149-1G, wherein the one in the multiple configuration mode refers to The non-null set for showing coefficient includes corresponding in the coefficient to be greater than one or more of the multiple spherical harmonic coefficient institute Those of the order of order of corresponding basis function coefficient, and exclude humorous corresponding to the multiple ball is greater than in the coefficient At least one coefficient of the order of the order of the corresponding basis function of the one or more in coefficient.

Clause 133149-5G.The device of clause 133149-1G, wherein the one in the multiple configuration mode refers to The non-null set for showing coefficient includes all coefficients, except at least one of described coefficient.

Clause 133149-6G.The device of clause 133149-1G, wherein one or more described processors are further configured With the multiple based on being determined in bit stream with the value that signal indicates when determining the one in the multiple configuration mode The one in configuration mode.

Figure 52 is to illustrate that the various aspects of implementable technology described in the present invention rebuild SHC 701 to rebuild or approach Audio decoding apparatus 750A another example block diagram.In the example of Figure 52, audio decoding apparatus 750A is similar to Figure 41 D Example shown in audio decoding apparatus 540D, extraction unit 542 receive bit stream 715'(its be similar to above in relation to Figure 51 Example described in bit stream 715, bit stream 715' also includes SHCBG752 through audio coding version except) and side channel letter Except breath 719.For this reason, the extraction unit is expressed as " extraction unit 542' ".

In addition, extraction unit 542' is different from being that extraction unit 542' includes V decompression unit in place of extraction unit 542 555 modified form (it is shown as " V decompression unit 555' " in the example of Figure 52).V decompression unit 555' is received Side channel information 719 and the syntactic element for being expressed as codedVVecLength 754.Extraction unit 542' is from bit stream 715' (and in an example, from the access unit header being contained in bit stream 715') dissects codedVVecLength 754.V solution Compression unit 555' includes mode configuration unit 756 (" mode configuration unit 756 ") and is configurable to according to before retouching configuration mode The anatomy unit 758 of any one of 760 operations.

Mode configuration unit 756 receives one of syntactic element 754 and option and installment mode 7 60.Mode configuration unit 756 then configure anatomy unit 758 with the selected person in configuration mode 760.It dissects the expression of unit 758 and is configured to basis Any one of configuration mode 760 is operated to dissect the unit of the compressed format of small [V] vector 717 from side channel information 719. Dissecting unit 758 can operate according to the switching sentence presented in following table.

The decoded vector of table-

In previous syntax table, there are four types of the first switching sentences of situation (situation 0 to 3) to provide to according to coefficient for tool Number determines the mode of the length of each vector of small [V] matrix 717.First situation, situation 0, instruction are used for VT DISTVector All coefficients are designated.Second situation, situation 1 indicate only VT DISTCorresponding to for vector is greater than Those of order of MinNumOfCoeffsForAmbHOA coefficient is designated, and can indicate to be referred to as (N aboveDIST+1)-(NBG+ 1) the case where.Third situation, situation 2 are similar to the second situation, but further subtract the coefficient of identification NumOfAddAmbHoaChan indicates to correspond to over order N for specifiedBGOrder additional channels (wherein " channel " Refer to correspond to a certain order, sub- order combinations particular factor) variable.4th situation, situation 3 indicate only VT DISTVector Those of remaining coefficient is designated after removing the coefficient identified by NumOfAddAmbHoaChan.

In in this respect, audio decoding apparatus 750A can according to the present invention described in technical operation indicate sound to determine First non-null set of the coefficient of the vector of the distinct components of field, the vector is from multiple spherical harmonic coefficients point of description sound field Solution.

In addition, audio decoding apparatus 750A can be configured it is multiple with determination with the technical operation described according to the present invention One of configuration mode, it is described by according to the non-null set of the coefficient of the one extracted vector in multiple configuration modes Vector is decomposed from multiple spherical harmonic coefficients of the distinct components of description sound field and expression sound field, and is based on the multiple configuration mode In the winner extract the vector coefficient the non-null set.

Figure 53 is the another of the audio coding apparatus 570 for the various aspects that technology described in the present invention can be performed in explanation The block diagram of example.In the example of Figure 53, audio coding apparatus 570 can be similar to one in audio coding apparatus 510A to 510J Or more persons are (wherein it is assumed that Degree reduction unit 528A is included in sound field component extraction unit 20 but to be easy the purpose of explanation not Diagram).However, audio coding apparatus 570 may include that more typically converter unit 572, the converter unit in some instances may be used Including decomposition unit 518.

Figure 54 is the frame that the example implementation of audio coding apparatus 570 shown in the example of Figure 53 is described in more detail Figure.As illustrated in the example of Figure 54, the converter unit 572 of audio coding apparatus 570 includes rotary unit 654.Audio coding The sound field component extraction unit 520 of device 570 includes spatial analysis unit 650, content character analytical unit 652, extracts and be concerned with Component cells 656 and extraction diffusion component unit 658.The audio coding unit 514 of audio coding apparatus 570 is decoded comprising AAC Engine 660 and AAC decode engine 162.The bitstream producing unit 516 of audio coding apparatus 570 includes multiplexer (MUX) 164。

It indicates it can be made to use in consumer according to the bandwidth of bit/second required for the 3D audio data in the form of SHC Aspect is excessively high.For example, when using the sampling rate of 48kHz, and in the case where 32/equal resolution, four Rank SHC expression is the bandwidth (25 × 48000 × 32bps) for indicating 36Mbits/ seconds.When with the current skill for stereo signal When art level audio decoding (it is usually about 100kbits/ seconds) compares, this is biggish number.Institute is real in the example of Figure 54 The technology applied can reduce the bandwidth of 3D audio representation.

Spatial analysis unit 650, content character analytical unit 652 and rotary unit 654 can receive SHC 511.Such as at this In invention described by other places, SHC 511 can indicate sound field.In the example of Figure 54, spatial analysis unit 650, content are special Property analytical unit 652 and rotary unit 654 can receive for sound field quadravalence (n=4) expression 25 SHC.

Spatial analysis unit 650 can analyze the sound field that is indicated by SHC 511 to identify the distinct components and sound field of sound field Diffusion component.The distinct components of sound field be perceived as from recognizable direction or in other ways different from the background of sound field or The sound of diffusion component.For example, it can be perceived as coming from identifiable direction by the sound that individual instruments generate.Compared to it Under, the diffusion of sound field or background component are not perceived as from identifiable direction.For example, the sound that wind passes through forest can For the diffusion component of sound field.

Spatial analysis unit 650 can recognize one or more distinct components for attempting to identify best angle, with the optimum angle Degree rotation sound field is so as to have those of the distinct components of most of energy component with vertical and/or trunnion axis (relative to record The hypothesis microphone of this sound field) alignment.Spatial analysis unit 650 can recognize this best angle so that sound field is rotatable, so that These distinct components are preferably aligned with spherical surface basis function in basis shown in the example of Fig. 1 and 2.

In some instances, spatial analysis unit 650 can indicate the diffusion analysis for being configured to execute a certain form to know (diffusion sound can be referred to low-level direction or low the percentage for the sound field indicated by SHC 511 that sound Bao Han not spread The sound of rank SHC, it is meant that there is those of the order less than or equal to one SHC 511).As an example, spatial analysis Unit 650 can be similar to the entitled " spatial sound decoded with directional audio of Willie pa lucky (Ville Pulkki) The paper of regeneration (Spatial Sound Reproduction with Directional Audio Coding) " (is published in Sense of hearing engineering science association proceedings the 6th phase of volume 55, the date be in June, 2007) described in mode mode execute diffusion point Analysis.In some cases, spatial analysis unit 650 can only analyze HOA system when executing diffusion analysis to determine diffused percentage Several non-zero subsets, such as the zeroth order and single order SHC of SHC 511.

Content character analytical unit 652 can be at least partially based on SHC 511 determine the SHC 511 be via sound field from So record, which generates, still artificially (that is, synthetically) is generated by the audio object of (as an example) such as PCM object.This Outside, content character analytical unit 652 can then be at least partially based on SHC 511 be via sound field physical record generate or by Artificial audio object generates the sum for determining the channel for including in bit stream 517.For example, content character analytical unit 652 can Being at least partially based on SHC 511 is to be generated by the record of practical sound field or generated by artificial audio object to determine bit stream 517 It will include 16 channels.Each of channel can be monophonic channel.Content character analytical unit 652 can be based on bit stream 517 output bit rate (for example, 1.2Mbps) further executes the determination of the sum in channel included in decode bit stream 517.

In addition, content character analytical unit 652, which can be at least partially based on SHC 511, to be generated also by the record of practical sound field It is to be generated by artificial audio object to determine that relevant or (in other words) distinct components of sound field are distributed in how many channel, and have Distribute to the diffusion or (in other words) background component of sound field in how many channel.For example, when SHC 511 be by use (as One example) Eigenmic record the generation of practical sound field when, content character analytical unit 652 can be by three channels in channel The coherent component of sound field is distributed to, and remaining channel can be distributed to the diffusion component of sound field.In this example, when by artificial sound When frequency object generates SHC 511, five channels in channel can be distributed to relevant point of sound field by content character analytical unit 652 Amount, and remaining channel can be distributed to the diffusion component of sound field.By this method, content analysis block is (that is, content character analytical unit 652) it can determine the type (for example, diffusion/direction etc.) of sound field, and determine the number for the relevant/diffusion component extracted again.

Targeted bit rates can influence the number of the component of individual AAC decoding engines (for example, AAC decodes engine 660,662) And bit rate.In other words, content character analytical unit 652 can further execute based on bit stream 517 output bit rate (such as It is 1.2Mbps) determining to distribute to the number of active lanes of coherent component and distribute to the number of active lanes of diffusion component.

In some instances, the bit rate for distributing to the channel of the coherent component of sound field can be greater than the diffusion for distributing to sound field The bit rate in the channel of component.For example, the maximum bitrate of bit stream 517 can be 1.2Mb/ seconds.In this example, it may be present It distributes to four channels of coherent component and distributes to 16 channels of diffusion component.In addition, in this example, distributing to relevant Each of channel of component can have 64kb/ seconds maximum bitrates.In this example, the channel of diffusion component is distributed to Each of can have 48kb/ seconds maximum bitrates.

As indicated above, content character analytical unit 652 can determine SHC 511 be from the record of practical sound field or from Artificial audio object generates.Content character analytical unit 652 can carry out this determination in various ways.For example, audio coding 4 rank SHC can be used in device 570.In this example, 652 decodable code of content character analytical unit, 24 channels and prediction the 25th Channel (it is represented by vector).Scalar can be applied at least one in 24 channels by content character analytical unit 652 It is added a bit and by resulting value to determine the 25th vector.In addition, in this example, content character analytical unit 652 can determine it is pre- Survey the accuracy in the 25th channel.In this example, if predicting that the accuracy in the 25th channel is relatively high (for example, accurate Degree is more than specific threshold), then SHC 511 may be to generate from Composite tone object.In contrast, if predicting the 25th channel Accuracy it is relatively low (for example, accuracy be lower than specific threshold), then SHC 511 is more likely to indicate recorded sound field.It lifts For example, in this example, if more than 100 decibels of the signal-to-noise ratio (SNR) (db) in the 25th channel, SHC 511 are more likely to table Show the sound field generated from Composite tone object.In contrast, the SNR of the sound field recorded using intrinsic microphone can be arrived for 5db 20db.Thus, by between the sound field that the SHC 511 that practical directly record generates and generates from Composite tone object is indicated Obvious boundary may be present in SNR ratio.

In addition, content character analytical unit 652 can be at least partially based on SHC 511 be from the record of practical sound field or from Artificial audio object generates and selects the codebook for quantifying V vector.In other words, the sound field depending on being indicated by HOA coefficient It is record or synthesis, content character analytical unit 652 may be selected different codebooks and supply for quantifying V vector.

In some instances, content character analytical unit 652 repeatably determines that SHC 511 is the record from practical sound field Or it is generated from artificial audio object.In these some examples, repeating basis can be each frame.In other examples, content Specificity analysis unit 652 can be performed this and determine once.In addition, content character analytical unit 652 repeatably determines coherent component The total number of channels and distribution in channel and diffusion component channel.In these some examples, repeating basis can be each frame.Other In example, content character analytical unit 652 can be performed this and determine once.In some instances, content character analytical unit 652 can It is repeatedly selected for the codebook for quantifying V vector.In these some examples, repeating basis can be each frame.In other realities In example, content character analytical unit 652 can be performed this and determine once.

The rotation process of HOA coefficient can be performed in rotary unit 654.As discussed elsewhere in the present invention (for example, opposite In Figure 55 and 55B), execution rotation process can reduce bits number needed for indicating SHC 511.In some instances, single by rotation The pivot analysis that member 652 executes is the example of singular value decomposition (" SVD ") analysis.Principal component analysis (" PCA "), isolated component point Analysing (" ICA ") and card, Nan-La Wei transformation (" KLT ") is the relevant technologies that possible be applicable in suddenly.

In the example of Figure 54, extracts coherent component unit 656 and receive from rotary unit 654 through rotating SHC 511.This Outside, extract coherent component unit 656 from through rotation SHC 511 extract it is associated with the coherent component of sound field those through rotating SHC 511。

In addition, extracting coherent component unit 656 generates one or more coherent component channels.It is each in coherent component channel Person may include the different subsets through rotating SHC 511 associated with the coherence factor of sound field.In the example of Figure 54, phase is extracted Dry component cells 656 can produce 1 to 16 coherent component channels.Led to by the coherent component that extraction coherent component unit 656 generates The number in road can be determined by distributing to by content character analytical unit 652 number of active lanes of the coherent component of sound field.By mentioning The bit rate in the coherent component channel for taking coherent component unit 656 to generate can be determined by content character analytical unit 652.

Similarly, it in the example of Figure 54, extracts diffusion component unit 658 and receives from rotary unit 654 through rotating SHC 511.Those warps associated with the diffusion component of sound field are extracted from through rotation SHC 511 in addition, extracting diffusion component unit 658 Rotate SHC 511.

In addition, extracting diffusion component unit 658 generates one or more diffusion component channels.It is each in diffusion component channel Person may include the different subsets through rotating SHC 511 associated with the diffusion coefficient of sound field.In the example of Figure 54, extracts and expand Dissipating component cells 658 can produce 1 to 9 diffusion component channels.The diffusion component channel generated by extraction diffusion component unit 658 Number can be determined by distributing to by content character analytical unit 652 number of active lanes of the diffusion component of sound field.By extracting The bit rate in the diffusion component channel that diffusion component unit 658 generates can be determined by content character analytical unit 652.

In the example of Figure 54, AAC codec can be used to come to by extraction coherent component unit for AAC decoding unit 660 The 656 coherent component channels generated are encoded.Similarly, AAC decoding unit 662 can be used AAC codec come to by mentioning The diffusion component channel for taking diffusion component unit 658 to generate is encoded.Multiplexer 664 (" multiplexer 664 ") can To encoded coherent component channel and encoded diffusion component channel together with side data (for example, true by spatial analysis unit 650 Fixed best angle) it is multiplexed to generate bit stream 517.

By this method, the technology can make audio coding apparatus 570 can determine indicate sound field spherical harmonic coefficient whether be It is generated from Composite tone object.

In some instances, audio coding apparatus 570 can based on spherical harmonic coefficient whether be from Composite tone object generate and Determine the subset for indicating the spherical harmonic coefficient of distinct components of sound field.In these and other example, audio coding apparatus 570 can be produced Raw bit stream is with the subset comprising spherical harmonic coefficient.In some cases, audio coding apparatus 570 can be to the subset of spherical harmonic coefficient Carry out audio coding, and generate bit stream with comprising spherical harmonic coefficient through audio coding subset.

In some instances, audio coding apparatus 570 can based on spherical harmonic coefficient whether be from Composite tone object generate and Determine the subset for indicating the spherical harmonic coefficient of background component of sound field.In these and other example, audio coding apparatus 570 can be produced Raw bit stream is with the subset comprising spherical harmonic coefficient.In these and other example, audio coding apparatus 570 can be to spherical harmonic coefficient The subset carry out audio coding, and generate bit stream with comprising spherical harmonic coefficient through audio coding subset.

In some instances, audio coding apparatus 570 can execute spatial analysis relative to spherical harmonic coefficient and be rotated with identifying By the angle for the sound field that spherical harmonic coefficient indicates, and rotation process is executed so that sound field is rotated identified angle to generate through screw Humorous coefficient.

In some instances, audio coding apparatus 570 can based on spherical harmonic coefficient whether be from Composite tone object generate and It determines the first subset for indicating the spherical harmonic coefficient of distinct components of sound field, and whether is from Composite tone object based on spherical harmonic coefficient It generates and determines the second subset for indicating the spherical harmonic coefficient of background component of sound field.In these and other example, audio coding Device 570 can be by the targeted bit rates pair higher than the targeted bit rates to the second main body progress audio coding to spherical harmonic coefficient First subset of spherical harmonic coefficient carries out audio coding.

By this method, the various aspects of the technology can enable audio coding apparatus 570 determine SCH according to following clause Whether 511 be to generate from Composite tone object.

Clause 132512-1.A kind of device, such as audio coding apparatus 570 comprising: wherein one or more described processing Device, which is further configured to determine, indicates whether the spherical harmonic coefficient of sound field is to generate from Composite tone object.

Clause 132512-2.The device of clause 132512-1, wherein one or more described processors are further configured to The spherical harmonic coefficient of sound field is indicated from storage when determining whether the spherical harmonic coefficient for indicating sound field is to generate from Composite tone object At least part of frame formula spherical harmonic coefficient matrix excludes primary vector to obtain through downscaled frame formula spherical harmonic coefficient matrix.

Clause 132512-3.The device of clause 132512-1, wherein one or more described processors are further configured to The spherical harmonic coefficient of sound field is indicated from storage when determining whether the spherical harmonic coefficient for indicating sound field is to generate from Composite tone object At least part of frame formula spherical harmonic coefficient matrix excludes primary vector to obtain through downscaled frame formula spherical harmonic coefficient matrix, and is based on institute It states the residual vector through downscaled frame formula spherical harmonic coefficient matrix and predicts the vector through downscaled frame formula spherical harmonic coefficient matrix.

Clause 132512-4.The device of clause 132512-1, wherein one or more described processors are further configured to The spherical harmonic coefficient of sound field is indicated from storage when determining whether the spherical harmonic coefficient for indicating sound field is to generate from Composite tone object At least part of frame formula spherical harmonic coefficient matrix excludes primary vector to obtain through downscaled frame formula spherical harmonic coefficient matrix, and at least portion Divide summation prediction of the ground based on the residual vector through downscaled frame formula spherical harmonic coefficient matrix described through downscaled frame formula spherical harmonic coefficient The vector of matrix.

Clause 132512-5.The device of clause 132512-1, wherein one or more described processors are further configured to Storage spherical harmonic coefficient is based at least partially on when determining whether the spherical harmonic coefficient for indicating sound field is to generate from Composite tone object The summation of residual vector of at least part of frame formula spherical harmonic coefficient matrix predict the vector of the frame formula spherical harmonic coefficient matrix.

Clause 132512-6.The device of clause 132512-1, wherein one or more described processors are further configured to Storage spherical harmonic coefficient is based at least partially on when determining whether the spherical harmonic coefficient for indicating sound field is to generate from Composite tone object The summation of residual vector of at least part of frame formula spherical harmonic coefficient matrix predict the vector of the frame formula spherical harmonic coefficient matrix, And error is calculated based on institute's predicted vector.

Clause 132512-7.The device of clause 132512-1, wherein one or more described processors are further configured to Storage spherical harmonic coefficient is based at least partially on when determining whether the spherical harmonic coefficient for indicating sound field is to generate from Composite tone object The summation of residual vector of at least part of frame formula spherical harmonic coefficient matrix predict the vector of the frame formula spherical harmonic coefficient matrix, And error is calculated based on the correspondence vector of institute's predicted vector and the frame formula spherical harmonic coefficient matrix.

Clause 132512-8.The device of clause 132512-1, wherein one or more described processors are further configured to Storage spherical harmonic coefficient is based at least partially on when determining whether the spherical harmonic coefficient for indicating sound field is to generate from Composite tone object The summation of residual vector of at least part of frame formula spherical harmonic coefficient matrix predict the vector of the frame formula spherical harmonic coefficient matrix, It and by error calculation is the summation of institute's predicted vector and the absolute value of the difference of the corresponding vector of the frame formula spherical harmonic coefficient matrix.

Clause 132512-9.The device of clause 132512-1, wherein one or more described processors are further configured to Storage spherical harmonic coefficient is based at least partially on when determining whether the spherical harmonic coefficient for indicating sound field is to generate from Composite tone object The summation of residual vector of at least part of frame formula spherical harmonic coefficient matrix predict the vector of the frame formula spherical harmonic coefficient matrix, Error is calculated based on the correspondence vector of institute's predicted vector and the frame formula spherical harmonic coefficient matrix, is based on the frame formula spherical harmonic coefficient square One ratio of energy and the error calculation of the correspondence vector of battle array, and the ratio and threshold value are to determine the ball for indicating sound field Whether humorous coefficient is to generate from Composite tone object.

Clause 132512-10.The device of any one of technical solution 4 to 9, wherein one or more described processors are further It is configured to the frame formula spherical harmonic coefficient matrix that the prediction when predicting the vector stores at least described part of the spherical harmonic coefficient The first non-vanishing vector.

Clause 132512-11.The device of any one of technical solution 1 to 10, wherein one or more described processors are into one Step is configured in the bit stream through compressed version of storage spherical harmonic coefficient and specifies whether spherical harmonic coefficient is from Composite tone object The instruction of generation.

Clause 132512-12.The device of clause 132512-11, wherein described be designated as single position.

Clause 132512-13.The device of clause 132512-1, wherein one or more described processors are further configured to It whether is to be generated from Composite tone object and determine the subset for indicating the spherical harmonic coefficient of distinct components of sound field based on spherical harmonic coefficient.

Clause 132512-14.The device of clause 132512-13, wherein one or more described processors are further configured To generate bit stream with the subset comprising spherical harmonic coefficient.

Clause 132512-15.The device of clause 132512-13, wherein one or more described processors are further configured Audio coding is carried out with the subset to spherical harmonic coefficient, and generates bit stream with sub through audio coding comprising the spherical harmonic coefficient Collection.

Clause 132512-16.The device of clause 132512-1, wherein one or more described processors are further configured to It whether is to be generated from Composite tone object and determine the subset for indicating the spherical harmonic coefficient of background component of sound field based on spherical harmonic coefficient.

Clause 132512-17.The device of clause 132512-16, wherein one or more described processors are further configured To generate bit stream with the subset comprising spherical harmonic coefficient.

Clause 132512-18.The device of clause 132512-15, wherein one or more described processors are further configured Audio coding is carried out with the subset to spherical harmonic coefficient, and generates bit stream with sub through audio coding comprising the spherical harmonic coefficient Collection.

Clause 132512-18.The device of clause 132512-1, wherein one or more described processors are further configured to Spatial analysis is executed to identify the angle that need to rotate the sound field indicated by spherical harmonic coefficient relative to spherical harmonic coefficient, and executes rotation behaviour Make so that the sound field is rotated identified angle to generate through rotating spherical harmonic coefficient.

Clause 132512-20.The device of clause 132512-1, wherein one or more described processors are further configured to It whether is to generate and determine the first of the spherical harmonic coefficient of the distinct components of expression sound field from Composite tone object based on spherical harmonic coefficient Subset, and whether be to be generated from Composite tone object and determine the spherical harmonic coefficient for indicating the background component of sound field based on spherical harmonic coefficient Second subset.

Clause 132512-21.The device of clause 132512-20, wherein one or more described processors are further configured To press the targeted bit rates higher than the targeted bit rates for carrying out audio coding to the second main body to spherical harmonic coefficient to the humorous system of ball The first several subsets carries out audio coding.

Clause 132512-22.The device of clause 132512-1, wherein one or more described processors are further configured to Singular value decomposition is executed relative to the spherical harmonic coefficient to generate the U square for the left singular vector for indicating the multiple spherical harmonic coefficient The V square of battle array, the s-matrix for indicating the multiple spherical harmonic coefficient singular value and the right singular vector for indicating the multiple spherical harmonic coefficient Battle array.

Clause 132512-23.The device of clause 132512-22, wherein one or more described processors are further configured Whether to be one or more of to be generated from Composite tone object and determine U matrix, s-matrix and V matrix based on spherical harmonic coefficient Indicate those of the distinct components of sound field part.

Clause 132512-24.The device of clause 132512-22, wherein one or more described processors are further configured Whether to be one or more of to be generated from Composite tone object and determine U matrix, s-matrix and V matrix based on spherical harmonic coefficient Indicate those of the background component of sound field part.

Clause 132512-1C.A kind of device, such as audio coding apparatus 570 comprising: one or more processors, it is described One or more processors are configured to the energy based on the vector depending at least on spherical harmonic coefficient and the vector based on spherical harmonic coefficient The vector of predicted version and spherical harmonic coefficient and derived error and calculated ratio determine indicate sound field spherical harmonic coefficient whether It is to be generated from Composite tone object.

In each of above-mentioned various examples, it should be appreciated that 570 executing method of audio coding apparatus, or in addition packet Include the device for executing each step that audio coding apparatus 570 is configured to the method executed.In some cases, these devices It may include one or more processors.In some cases, one or more described processors can be indicated by means of storage to nonvolatile Property computer-readable storage medium instruction configuration application specific processor.In other words, each of the set of encoding example In the various aspects of technology can provide the non-transitory computer-readable storage media for being stored thereon with instruction, described instruction exists One or more described processors are caused to execute the method that audio coding apparatus 570 has been configured to execute when execution.

Figure 55 and 55B is to illustrate to execute the various aspects of technology described in the present invention to rotate the example of sound field 640 Figure.Figure 55 be according to the present invention described in technology various aspects sound field 640 of the explanation before rotation figure.Scheming In 55 example, sound field 640 includes two high pressure positions (being expressed as position 642A and 642B).These positions 642A and 642B (" position 642 ") be located at non-zero slope (it be the another way with reference to non-aqueous horizontal line because horizontal line have zero it is oblique Rate) line 644 on.It is assumed that position 642 also has z coordinate in addition to x and y-coordinate, it may be necessary to higher-order spherical substrate function Correctly to indicate this sound field 640 (because these higher-order spherical substrate functions describe the top and lower part or non-horizontal of sound field Part).The rotatable sound field 640 of audio coding apparatus 570 is and indirect by sound until 644 level of line of link position 642 Field 640 is reduced to SHC 511.

Figure 55 B is the figure for illustrating sound field 640 after being rotated up 644 level of line of link position 642.Due to this Mode rotates sound field 640, makes the higher-order SHC in SHC 511 be designated as zero (it is assumed that through revolving so SHC 511 can be exported Turning sound field 640 no longer has z coordinate any pressure (or energy) position).By this method, audio coding apparatus 570 can revolve Turn, translate or adjust more by and large sound field 640 to reduce the number of the SHC 511 with nonzero value.In conjunction with the technology Various other aspects, audio coding apparatus 570 then can indicate that unused signal indicates SHC with signal in the field of bit stream 517 511 these higher-orders SHC, rather than indicate that these higher-orders SHC of identification SHC 511 has 32 bands of zero just with signal Negative sign number.Audio coding apparatus 570 also azimuth often can be expressed by mode described above and elevation angle is in place The rotation information of rotation sound field 640 is indicated how in stream 517.Such as the extraction elements such as audio coding apparatus can be implied then The SHC that these unused signals of SHC 511 indicate has zero, and when regenerating sound field 640 based on SHC 511, executes rotation To rotate sound field 640 so that sound field 640 shown in example of the sound field 640 similar to Figure 55.By this method, audio coding fills Set 570 can according to the present invention described in technology reduce need in bit stream 517 specify SHC 511 number.

' space compression ' algorithm can be used to determine the best rotation of sound field.In one embodiment, audio coding apparatus 570 the algorithm can be performed be iterating through all possible azimuth and elevation angle combination (that is, in the above example for 1024 × 512 combinations), to be directed to each combination rotation sound field and calculate the number for being higher than the SHC 511 of threshold value.It can will generate minimum The azimuth of the SHC 511 higher than threshold value of number/elevation angle candidate combination is considered as the combination that can be referred to as " best rotation ".Herein Through in rotated versions, sound field may need the SHC 511 of minimal amount with for indicate sound field and can thus be considered as compression. In some cases, adjustment may include this best rotation and adjustment information as described above may include that (it can be described as this rotation " best rotation ") information (for azimuth and the elevation angle).

In some cases, audio coding apparatus 570 form of Eulerian angles can come outside specified amount (as an example) Angle, rather than only specify azimuth and the elevation angle.The specified rotation angle about z-axis, pervious x-axis and pervious z-axis of Eulerian angles.Though So the combination in the present invention relative to azimuth and the elevation angle is described, but technology of the invention should not be so limited to only designated parties Parallactic angle and the elevation angle, but may include specifying any number of angle (including three Eulerian angles referred to above).In this meaning On, the rotatable sound field of audio coding apparatus 570 is to reduce the number of offer to the multiple stratum's elements for describing the relevant information of sound field Mesh and Eulerian angles are appointed as rotation information in bit stream.As mentioned above, Eulerian angles may describe how rotation sound field.When making When with Eulerian angles, bit stream extraction element can dissect bit stream to determine the rotation information for including Eulerian angles, and provide and retouch when be based on It states and rotates sound field based on Eulerian angles when the multiple stratum's members of those of relevant information of sound field usually regenerate sound field.

In addition, in some cases, audio coding apparatus 570 may specify predetermined with one or more angles of specified rotation The associated index of justice combination (it can be referred to as " rotation index "), rather than these angles are explicitly specified in bit stream 517.In other words It says, in some cases, rotation information may include rotation index.In these cases, the given value of index is rotated (for example, zero Value) it can indicate to be not carried out rotation.It can come to use this rotation index about rotation table.That is, audio coding apparatus 570 may include rotation Table, the rotation table include the entry for each of the combination at azimuth and the elevation angle.

Alternatively, rotation table may include the entry for each matrixing for each combination for indicating azimuth and the elevation angle. That is, audio coding apparatus 570 can store rotation table, the rotation table has for for by sound field rotational orientation angle and the elevation angle The entry of each matrixing of each combination in combination.In general, audio coding apparatus 570 receives SHC 511 and when execution SHC 511' is exported when rotation according to following equation:

In above equation, SHC 511' is calculated as to the function of following three: for compiling according to the second referential Encoder matrix (the EncMat of code sound field2);For SHC 511 to be restored back to the inverse matrix of the sound field according to the first referential (InvMat1);And SHC 511.EncMat2With size 25x32, and InvMat1With size 32x25.SHC 511' and SHC 511 both have size 25, and wherein SHC 511' is attributable to not removing specifiedly those of prominent audio-frequency information SHC and obtain Further to reduce.EncMat2It can change for each azimuth and elevation angle combination, and InvMat1It can be relative to each orientation Angle and the elevation angle are combined and are remained unchanged.Rotation table may include storing each difference EncMat2With InvMat1The result of multiplication Entry.

Figure 56 is the figure for illustrating the example sound field captured according to the first referential, and first referential is then according to this hair Technology rotation described in bright is to express sound field according to the second referential.In the example of Figure 56, it is assumed that the first referential In the case where capture surround the sound field of intrinsic microphone 646, first referential is in the example of Figure 56 by X1、Y1And Z1Axis It indicates.SHC 511 describes sound field according to this first referential.InvMat1SHC 511 is transformed back to sound field, thus scheming Make it possible to rotate to sound field by X in 56 example2、Y2And Z2Second referential represented by axis.It is as described above EncMat2Rotatable sound field simultaneously generates the second referential of foundation to describe this SHC 511' through rotating sound field.

Under any circumstance, above equation can be exported as follows.It is assumed that sound field is recorded with a certain coordinate system, so that front It is considered as the direction of x-axis, 32 microphones of intrinsic microphone (or the configuration of other microphones) is defined from this reference frame Position.The rotation of sound field can be then considered as to the rotation of this referential.For the referential assumed, SHC can be calculated as follows 511:

In above equation,Indicate the position (Pos of the i-th microphone (wherein i can be 1 to 32 in this example)i) The spherical substrate function at place.miciVector indicates the microphone signal of the i-th microphone of time t.Position (Posi) refer to that microphone exists Position in first referential (that is, being in this example the referential before rotation).

Alternatively above equation is expressed according to Mathematical representation represented above are as follows:

In order to rotate sound field (or in second referential), will in the second referential calculating position (Posi).As long as former Beginning microphone signal exists, and can arbitrarily rotate sound field.However, original microphone signal (mici(t)) often unavailable.It asks Topic then can for how from SHC 511 retrieve microphone signal (mici(t)).If using T design (such as in 32 microphones it is intrinsic In microphone), then this solution to the problem can be realized by solving following equation:

This InvMat1It may specify humorous according to position (as relative to specified by the first referential) ball calculated of microphone Basis function.This equation also can be expressed asAs described above.

Once retrieving microphone signal (mic according to above equationi(t)), can rotation description sound field the Mike Wind number (mici(t)) to calculate the SHC 511' for corresponding to the second referential, to generate following equation:

EncMat2It is specified to come from rotation position (Posi') the humorous basis function of ball.By this method, EncMat2It can be effectively The combination at specified azimuth and the elevation angle.Therefore, when rotation table stores each combination at azimuth and the elevation angleResult when, rotation table effectively specify azimuth and the elevation angle each combination.Above equation Also it can be expressed as:

Wherein θ2,It indicates to be different from by θ1,It the first party parallactic angle of expression and the second party parallactic angle at the elevation angle and second faces upward Angle.θ1,Corresponding to the first referential, and θ2,Corresponding to the second referential.InvMat1Can therefore it correspond toAnd EncMat2It can correspond to

It can indicate not considering filtering operation (above in the various equations for indicating to export SHC 511 in a frequency domain above By jn() function representation, the jn() function refers to n rank spherical surface Bessel function) calculating more simple version.In the time domain, This jnFiltering operation of the () function representation particularly for specific order n.In the case where being filtered, rotation can be executed by order Turn.To be illustrated, consideration following equation:

From these equations, dividually complete several orders through rotate SHC 511', because of the b for every single ordern (t) it is different.As a result, above equation can be changed as follows for calculating through the single order person in rotation SHC 511':

It is assumed that there are three single order SHC 511, each of 511 vector of SHC 511' and SHC in above equation Size be three.For second order, following equation can be applied:

Again, it is assumed that there are five second order SHC 511, in above equation in 511 vector of SHC 511' and SHC The size of each is five.For other ranks (that is, three ranks and quadravalence), residual equation formula can be similar to as described above Equation, the size about matrix follow model identical (because of EncMat2Line number, InvMat1Columns and three rank SHC The size of 511 and SHC 511' vector and quadravalence SHC 511 and SHC 511' vector are equal to the humorous basis function of three rank balls and quadravalence The number of the sub- rank of each of the humorous basis function of ball (m multiplies two and adds 1).

Therefore audio coding apparatus 570 can execute this rotation process relative to each combination at azimuth and the elevation angle to taste Examination identifies so-called best rotation.After executing this rotation process, audio coding apparatus 570 can calculate the SHC higher than threshold value The number of 511'.In some cases, audio coding apparatus 570 can execute this rotation within a duration (for example, audio frame) Turn to export a series of SHC 511' for indicating sound field.Sound field is indicated by executing this rotation within this duration to export A series of SHC 511', audio coding apparatus 570 can have to hold less than reduction in the duration of a frame or other length The number of capable rotation process (compared with completing this rotation process for each set of the SHC 511 of description sound field).It is in office In the case of what, audio coding apparatus 570 can save those SHC 511' through this process, to be greater than threshold with minimal amount The SHC 511' of value.

However, executed relative to each combination at azimuth and the elevation angle this rotation process can for processor it is intensive or consumption When.As a result, audio coding apparatus 570 can not execute the process for being characterized by this " brute-force " embodiment of Rotation Algorithm. Alternatively, audio coding apparatus 570 (can press statistics known to the substantially azimuth of good compression and the possibility at the elevation angle relative to providing Mode) subset of combination executes rotation, it is further rotated to execute about the combination around the combination in this subset, thus with Other combinations in subset are compared to offer preferably compression.

As another alternative, audio coding apparatus 570 can execute this rotation only with respect to combined known subset.Make For another alternative, audio coding apparatus 570 can follow the track (spatially) of combination, execute relative to this track of combination Rotation.As another alternative, audio coding apparatus 570 may specify compression threshold, and the compression threshold, which defines to have, is higher than threshold The maximum number of the SHC 511' of the nonzero value of value.This compression threshold can effectively set the halt of search, so that working as audio Code device 570 execute rotation and determine have higher than set threshold value value SHC 511' number be less than or equal to (or Be less than in some cases) compression threshold when, audio coding apparatus 570 stops executing any additional rotation relative to residue combination Turn operation.As another alternative, audio coding apparatus 570 can traverse class type arrangement tree (or other data knots of combination Structure), the number for executing rotation process and depending on the SHC 511' with the nonzero value greater than threshold value is closed relative to present combination Mesh and traverse the tree to the right or the left side (for example, for binary tree).

In this sense, each of these alternatives are related to executing first and second rotation process and compare execution One and second rotation process result to identify the of the SHC 511' with the nonzero value greater than threshold value for generating minimal amount One and second one of the rotation process.Therefore, audio coding apparatus 570 can execute the first rotation process according to the to sound field One azimuth and first elevation angle rotate sound field, and determine that the multiple class type for providing information relevant with description sound field is first First number of element, the sound field that the multiple stratum's element representation is rotated according to first party parallactic angle and first elevation angle.Audio is compiled Code device 570 can execute the second rotation process to sound field also to rotate sound field according to second party parallactic angle and second elevation angle, and really Second number of fixed multiple class type elements that information relevant to description sound field is provided, the multiple class type element representation root The sound field rotated according to second party parallactic angle and second elevation angle.In addition, audio coding apparatus 570 can be based on the multiple class type member First number of element selects the first rotation process or the second rotation compared with the second number of the multiple class type element Operation.

In some instances, Rotation Algorithm can be executed relative to the duration, wherein to the subsequent calls of Rotation Algorithm It can be called based on the past to Rotation Algorithm to execute rotation process.In other words, Rotation Algorithm can be based on going through in rotation sound field When previous duration when it is identified in the past rotation information and be adaptive.For example, audio coding apparatus 570 can revolve Turning sound field lasts the first duration (for example, audio frame) to identify the SHC 511' for being directed to this first duration.Audio is compiled Code device 570 can specify rotation information and SHC 511' any one of in a manner of as described above and in bit stream 517.It can This rotation information is referred to as the first rotation information, because it describes rotation of the sound field in the first duration.Audio coding dress Set 570 can then be rotated based on this first rotation information sound field last the second duration (for example, second audio frame) with know SHC 511' of the safety pin to this second duration.When executing the second rotation process within the second duration, audio coding Device 570 can be using the search of " best " combination of this first rotation information to initialize azimuthal and the elevation angle (as one Example).Audio coding apparatus 570 can then specify SHC 511' in bit stream 517 and revolve for the corresponding of the second duration Transfering the letter breath (it can be referred to as " the second rotation information ").

Although being to be subject to relative to implementation Rotation Algorithm with several different modes for reducing processing time and/or consumption above Description, but the technology can be relative to can reduce or accelerate in other ways the identification to the rotation that can be referred to as " best rotation " Any algorithm is executed.In addition, relative to the rotation of identification non-optimal but performance can be improved in other aspects (often according to speed Degree or processor or other resource utilizations measure) any algorithm execute the technology.

Figure 57 to 57E be respectively illustrate according to the present invention described in technology formed bit stream 517A to 517E figure. In the example of Figure 57 A, bit stream 517A can indicate an example of bit stream 517 shown in figure 5 above 3.Bit stream 517A includes SHC There are the fields of field 670 and storage SHC 511' (wherein the field list is shown as " SHC 511' ").There are fields 670 by SHC It may include the position corresponding to each of SHC 511.SHC 511' can indicate that referring in bit stream for SHC 511 is those of fixed SHC can be less than the number of SHC 511 in number.In general, each of SHC 511' is SHC 511 with non-zero Those of value SHC.It is indicated as described previously for the quadravalence of any given sound field, needs (1+4)2Or 25 SHC.Eliminate these One or more of SHC and it these zeros SHC is replaced with single position can save 31, it can be the allocated to express sound field in more detail Other parts or in addition remove to promote efficient bandwidth usage.

In the example of Figure 57 B, bit stream 517B can indicate an example of bit stream 517 shown in figure 5 above 3.Bit stream (wherein the field list is shown as field of the 517B comprising information converting field 672 (" information converting 672 ") and storage SHC 511' "SHC 511'").As described above, information converting 672 may include the adjustment of address translation information, rotation information and/or expression to sound field Any other form information.In some cases, information converting 672, which also may specify, is appointed as SHC in bit stream 517B The top step number of the SHC 511 of 511'.That is, information converting 672 can indicate that three order, extraction element can be regarded as instruction SHC 511' includes up to and includes those of those of order SHC 511 SHC 511 with three.Extraction element can be then configured Will have the SHC 511 of four or higher order number to be set to zero, and then it is four or higher that order is potentially removed in bit stream The explicit signaling of SHC 511.

In the example of Figure 57 C, bit stream 517C can indicate an example of bit stream 517 shown in figure 5 above 3.Bit stream 517C includes information converting field 672 (" information converting 672 "), there are field 670 and the fields of storage SHC 511' by SHC (wherein the field list is shown as " SHC 511' ").It is not configured to understand as above in relation to not having to letter described by Figure 57 B Which order of SHC 511 number indicated, there are fields 670 explicitly to indicate which one of SHC 511 flows in place with signal by SHC SHC 511' is appointed as in 517C.

In the example of Figure 57 D, bit stream 517D can indicate an example of bit stream 517 shown in figure 5 above 3.Bit stream 517D includes order field 674 (" order 60 "), there are fields 670, azimuth flag 676 (" AZF 676 "), elevation angle flag by SHC Mark 678 (" ELF 678 "), azimuth field 680 (" azimuth 680 "), elevation angle field 682 (" elevation angle 682 ") and storage SHC The field of 511' (field described in again in which is expressed as " SHC 511' ").Order field 674 specifies the order of SHC 511', That is, the order for being indicated to the top step number for indicating the spherical substrate function of sound field by n above.Order field 674 is through opening up 8 bit fields are shown as, but there can be various other sizes, such as three (it is the number of position needed for specified quadravalence).SHC exists Field 670 is shown as 25 bit fields.However, there are fields 670 can have various other sizes by SHC again.There are fields by SHC 670 are shown as 25 to indicate that SHC there are field 670 may include for corresponding in the spherical harmonic coefficient that indicates of quadravalence of sound field One position of each.

Azimuth flag 676 indicates 1 flag, and azimuth field 680 is specified to whether there is in bit stream 517D.The side of working as Parallactic angle flag 676 is set to for the moment, and the azimuth field 680 of SHC 511' is present in bit stream 517D.When azimuth flag 676 When being set to zero, the azimuth field 680 of SHC 511' is not present or in addition specifies in bit stream 517D.Equally, elevation angle flag 678 indicate a flag, and elevation angle field 682 is specified to whether there is in bit stream 517D.When elevation angle flag 678 is set to one When, the elevation angle field 682 of SHC 511' is present in bit stream 517D.When elevation angle flag 678 is set to zero, SHC 511''s is faced upward Angle field 682 is not present or in addition specifies in bit stream 517D.Although being described as a signaling existing for corresponding field and corresponding word Zero signaling that section is not present, but convention is reversible so that zero specified corresponding field is specified in bit stream 517D and a specified correspondence Field is specified not in bit stream 517D.Technology described in the present invention therefore should not in this regard in be restricted.

Azimuth field 680 indicates 10 bit fields, the specified azimuth in the presence of in bit stream 517D.Although being shown as 10 bit fields, but azimuth field 680 can have other sizes.Elevation angle field 682 indicates 9 bit fields, when in bit stream The elevation angle is specified in the presence of in 517D.The azimuth and the elevation angle specified in field 680 and 682 respectively can be with the knots of flag 676 and 678 Closing indicates above-mentioned rotation information.This rotation information can be used to rotate sound field to restore the SHC 511 in original reference system.

SHC 511' field is shown as the variable field with size X.SHC 511' field is attributable in bit stream The number (as there are represented by field 670 by SHC) of appointed SHC 511' and change.Size X can be exported as SHC and be deposited One number in field 670 multiplies the function of 32 (it is the size of every SHC 511').

In the example of Figure 57 E, bit stream 517E can indicate another example of bit stream 517 shown in figure 5 above 3.Bit stream 517E includes order field 674 (" order 60 "), there are fields 670 and rotation index field 684, and storage SHC by SHC The field of 511' (field described in again in which is expressed as " SHC 511' ").There are field 670 and SHC by order field 674, SHC 511' field can be substantially similar to the above field.Rotation index field 684 can be indicated to the specified elevation angle and orientation 20 bit fields of one of a combination in 1024 × 512 (or in other words 524288) at angle.In some cases, only 19 It can be used to specify this to rotate index field 684, and audio coding apparatus 570 can refer in bit stream and determine additional flag to indicate to rotate Whether operation executes (and therefore, rotating index field 684 whether there is in bit stream).This rotation index field 684 is specified Rotation mentioned above index can be referred to common to both audio coding apparatus 570 and bit stream extraction element in rotation table Entry.This rotation table can store the various combination at azimuth and the elevation angle in some cases.Alternatively, rotation table can store above-mentioned Matrix effectively stores the various combination at azimuth and the elevation angle in the matrix form.

Figure 58 is that audio coding apparatus 570 shown in the example of explanatory diagram 53 implements technology described in the present invention Rotate the flow chart of the example operation of aspect.Initially, audio coding apparatus 570 can be according to one in above-mentioned various Rotation Algorithms Or more persons select azimuth and elevation angle combination (800).Audio coding apparatus 570 then can be rotated according to selected azimuth and the elevation angle Sound field (802).As described above, audio coding apparatus 570 can use InvMat mentioned above first1It is led from SHC 511 Sound field out.Audio coding apparatus 570, which may further determine that, indicates the SHC 511'(804 through rotating sound field).Although being described as individually Step or operation, but audio coding apparatus 570 can (it can be indicated using the transformation of selection for indicating azimuth and elevation angle combination [EncMat2][InvMat1] result), export sound field from SHC 511, rotating the sound field and determining indicates through rotation sound field SHC 511'。

Under any circumstance, audio coding apparatus 570 can then calculate the number of the identified SHC 511' greater than threshold value Mesh, this number is compared with for the previous ones number calculated combined relative to prior orientation angle and the elevation angle (806, 808).In relative to first party parallactic angle and the first iteration of elevation angle combination, this compares can be for (it can with predefined previous number It is set to comparison zero).Under any circumstance, if the identified number of SHC 511' is less than previous number ("Yes" 808), So audio coding apparatus 570 stores SHC 511', azimuth and the elevation angle, usually replaces and stores from the previous ones of Rotation Algorithm Previous SHC 511', azimuth and the elevation angle (810).

If the identified number of SHC 511' replaces previously having been deposited not less than previous number ("No" 808) or in storage After SHC 511' of storage, azimuth and SHC 511' at the elevation angle, azimuth and the elevation angle, audio coding apparatus 570 can determine rotation Turn whether algorithm is completed (812).That is, audio coding apparatus 570 can be determined whether evaluated azimuth as an example With whole available combinations at the elevation angle.In other examples, audio coding apparatus 570 can be determined whether to meet other criterion (such as The defined subset of whole of combination has been executed, given trace whether has been traversed, whether has traversed class type tree and arrive leaf node etc.) So that audio coding apparatus 570 is completed to execute Rotation Algorithm.If do not completed ("No" 812), audio coding apparatus 570 Above procedure (800-812) can be executed relative to another selected combination.If completed ("Yes" 812), audio coding apparatus 570 can by it is above-mentioned it is various in a manner of one of stored SHC 511', azimuth and the elevation angle (814) are specified in bit stream 517.

Figure 59 is that audio coding apparatus 570 shown in the example of explanatory diagram 53 executes technology described in the present invention Convert the flow chart of the example operation of aspect.Initially, audio coding apparatus 570 may be selected to indicate the matrix of Linear Invertible Transforms (820).Indicate the matrix that an example of the matrix of Linear Invertible Transforms can show for top, i.e. [EncMat1][IncMat1] Result.Audio coding apparatus 570 then can by the matrix application in sound field to convert the sound field (822).Audio coding Device 570, which may further determine that, indicates the SHC 511'(824 through rotating sound field).Although being described as independent step or operation, sound Frequency code device 570 can (it can indicate [EncMat using transformation2][InvMat1] result), from SHC 511 export sound field, become It changes the sound field and determines the SHC 511' for indicating transformation sound field.

Under any circumstance, audio coding apparatus 570 can then calculate the number of the identified SHC 511' greater than threshold value Mesh, so that this number is compared with the previous ones number calculated for the preceding application relative to transformation matrix (826, 828).If the identified number of SHC 511' is less than previous number ("Yes" 828), audio coding apparatus 570 is stored SHC 511' and matrix (or its a certain derivative, such as the index with matrix correlation connection), often replacement is from the previous of Rotation Algorithm Previous SHC 511' and matrix that iteration is stored (or derivatives thereof) (830).

If the identified number of SHC 511' is not less than previous number ("No" 828) or in storage SHC 511' and square After battle array replaces the SHC 511' and matrix that had previously been stored, audio coding apparatus 570 can determine whether transformation algorithm is completed (832).That is, audio coding apparatus 570 can be determined whether evaluated can all use transformation matrix as an example.Other In example, audio coding apparatus 570 can be determined whether to meet other criterion and (such as execute and can use whole warps of transformation matrix Subset is defined, given trace whether has been traversed, whether has traversed class type tree and arrive leaf node etc.) so that audio coding apparatus 570 complete to hold row-action method.If do not completed ("No" 832), audio coding apparatus 570 can be relative to another selected change It changes matrix and executes above procedure (820 to 832).If completed ("Yes" 832), audio coding apparatus 570 can be with above-mentioned each One of kind mode specifies stored SHC 511' and matrix (834) in bit stream 517.

In some instances, single iteration can be performed in transformation algorithm, assesses single transformation matrix.That is, transformation matrix can wrap Include any matrix for indicating Linear Invertible Transforms.In some cases, Linear Invertible Transforms can transform from a spatial domain to sound field Frequency domain.The example of this Linear Invertible Transforms may include Discrete Fourier Transform (DFT).The application of DFT can only relate to single iteration, It and therefore will not necessarily include that the step of whether transformation algorithm is completed determined.Therefore, the technology should not necessarily be limited by the reality of Figure 59 Example.

In other words, an example of Linear Invertible Transforms is Discrete Fourier Transform (DFT).It can be by DFT to 20 Five SHC 511' are operated to form the set of 25 recombination coefficients.Audio coding apparatus 570 can also be by described 25 The integer multiple that a SHC 511' zero filling is 2, so as to potentially increase DFT section size resolution ratio, and potentially have The more effective embodiment of DFT, such as by applying Fast Fourier Transform (FFT).In some cases, increase point of DFT Resolution is not necessarily needed beyond 25 points.In the transform domain as illustrated, audio coding apparatus 570 can threshold application with determine in given zone Between in whether there is any spectrum energy.Audio coding apparatus 570 then can abandon in this context or pulverised is lower than this threshold value Spectral coefficient energy, and audio coding apparatus 570 can restore to have abandoned using inverse transformation or pulverised SHC 511' in one or The SHC 511' of more persons.That is, the coefficient lower than threshold value is not present, and therefore after application inverse transformation, less bits can be used Sound field is encoded.

In one or more examples, described function can be implemented with hardware, software, firmware, or any combination thereof.Such as Fruit is implemented with software, then the function can be used as one or more instructions or codes be stored on computer-readable media or It is transmitted via computer-readable media, and is executed by hardware based processing unit.Computer-readable media may include Computer-readable storage medium corresponds to tangible medium, such as data storage medium, or promotes comprising any by computer journey Sequence is transmitted to the communication medium of the media (for example, according to communication protocol) at another place from one.By this method, computer-readable matchmaker Body generally can correspond to (1) tangible computer readable memory medium, be nonvolatile shape, or (2) communication medium, such as believe Number or carrier wave.Data storage medium can be can be by one or more computers or the access of one or more processors to retrieve for implementing Any available media of technical instruction described in the present invention, code and/or data structure.Computer program product can wrap Containing computer-readable media.

By way of example and not limitation, such computer-readable storage medium may include RAM, ROM, EEPROM, CD-ROM Or it other optical disk storage apparatus, disk storage device or other magnetic storage devices, flash memory or any other can be used to The expectation program code and media accessible by a computer of the form of store instruction or data structure.Equally, any connection It can properly be referred to as computer-readable media.For example, if using coaxial cable, Connectorized fiber optic cabling, twisted pair, digital subscriber Line (DSL) or the wireless technology such as infrared ray, radio and microwave are transmitted from website, server or other remote sources to be instructed, Coaxial cable, Connectorized fiber optic cabling, twisted pair, DSL or the wireless technology such as infrared ray, radio and microwave are so included in matchmaker In the definition of body.However, it is understood that computer-readable storage medium and data storage medium and do not include connection, carrier wave, signal Or other temporary media, but actually it is directed to the tangible storage medium of non-transitory.As used herein, disk and CD Comprising compact disk (CD), laser-optical disk, optical compact disks, digital versatile disc (DVD), floppy discs and Blu-ray Disc, Middle disk usually magnetically reproduce data, and CD with laser reproduce data optically.The combination of above-mentioned items It should be included in the range of computer-readable media.

Instruction can be executed by one or more processors, one or more described processors are, for example, at one or more digital signals Manage device (DSP), general purpose microprocessor, specific integrated circuit (ASIC), Field Programmable Logic Array (FPGA) or other equivalent Integrated or discrete logic.Therefore, " processor " can refer to aforementioned structure or be suitable for reality as used herein, the term Apply any one of any other structure of technology described herein.In addition, in certain aspects, function described herein Energy property can be provided in be configured in the specialized hardware and/or software module of encoding and decoding, or be incorporated into group and compile in collaboration with In decoder.Also, the technology can be fully implemented in one or more circuits or logic elements.

Technology of the invention can be implemented in a wide variety of devices or devices, include wireless handset, integrated circuit (IC) or one group of IC (for example, chipset).It is to emphasize to be configured to hold that various components, modules, or units are described in the present invention In terms of the function of the device of row disclosed technology, but not necessarily need to pass different hardware unit realization.In fact, as retouched above It states, various units can cooperate suitable software and/or firmware combinations in a codec hardware unit, or by mutual The set of hardware cell is operated to provide, the hardware cell includes one or more processors as described above.

The various embodiments of the technology have been described.The these and other aspects of the technology are in the appended claims In range.

Claims (60)

1. a kind of method for handling audio data comprising:
The the first non-null set for indicating the coefficient of the vector of distinct components of sound field is obtained, the vector is from the description sound field Multiple spherical harmonic coefficients linearly decompose;
One of multiple configuration modes are obtained, so as to extracting the vector according to the one in the multiple configuration mode Coefficient the non-null set, wherein the one in the multiple configuration mode indicates the non-null set of the coefficient Closing includes all coefficients;And
The non-null set of the coefficient of the vector is extracted based on one of the acquisition in the multiple configuration mode It closes.
2. according to the method described in claim 1, it further comprises the described first non-null set conduct for extracting the coefficient The first part of the vector.
3. according to the method described in claim 1, further comprising:
The described first non-null set of the vector is extracted from side channel information;And
Described first non-null set of the coefficient based on the vector obtain the multiple spherical harmonic coefficient through reassembling into version This.
4. according to the method described in claim 1, wherein the vector includes using linear decomposition from the multiple spherical harmonic coefficient The vector linearly decomposed.
5. according to the method described in claim 4, wherein the linear decomposition includes singular value decomposition.
6. according to the method described in claim 1, wherein the described first non-null set of coefficient is indicate the first sound field first Described first non-null set of the coefficient of the primary vector of distinct components, the method further includes:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include corresponding in the coefficient to be greater than one or more of more than described second a spherical harmonic coefficients Those of corresponding order of order of basis function coefficient;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
7. according to the method described in claim 1, wherein the described first non-null set of coefficient is indicate the first sound field first Described first non-null set of the coefficient of the primary vector of distinct components, the method further includes:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include corresponding in the coefficient to be greater than one or more of more than described second a spherical harmonic coefficients Those of corresponding order of order of basis function coefficient, and exclude corresponding to greater than more than described second in the coefficient At least one coefficient of the order of the order of the basis function corresponding to the one or more in a spherical harmonic coefficient; And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
8. according to the method described in claim 1, wherein the described first non-null set of coefficient is indicate the first sound field first Described first non-null set of the coefficient of the primary vector of distinct components, the method further includes:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include all coefficients, except at least one of described coefficient;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
9. according to the method described in claim 1, wherein obtaining the one in the multiple configuration mode includes based on position The one in the multiple configuration mode is obtained with the value that signal indicates in stream.
10. a kind of for handling the device of audio data comprising:
One or more processors, are configured to:
The the first non-null set for indicating the coefficient of the vector of distinct components of sound field is obtained, the vector is from the description sound field Multiple spherical harmonic coefficients linearly decompose;
One of multiple configuration modes are obtained, so as to extracting the vector according to the one in the multiple configuration mode Coefficient the non-null set, wherein the one in the multiple configuration mode indicates the non-null set of the coefficient Closing includes all coefficients;And
The non-null set of the coefficient of the vector is extracted based on one of the acquisition in the multiple configuration mode It closes.
11. device according to claim 10, wherein one or more described processors are further configured to described in extraction First part of the described first non-null set of coefficient as the vector.
12. device according to claim 10, wherein one or more described processors are further configured to lead to from side Described first non-null set of vector described in road information extraction, and the described first non-null set of the coefficient based on the vector Close obtain the multiple spherical harmonic coefficient through reassembling into version.
13. device according to claim 10, wherein the vector includes using linear decomposition from the humorous system of the multiple ball The vector that number linearly decomposes.
14. device according to claim 13, wherein the linear decomposition includes singular value decomposition.
15. device according to claim 10, wherein the described first non-null set of coefficient is indicate the first sound field the Described first non-null set of the coefficient of the primary vector of one distinct components, wherein one or more described processors are further through matching Set with:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include corresponding in the coefficient to be greater than one or more of more than described second a spherical harmonic coefficients Those of corresponding order of order of basis function coefficient;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
16. device according to claim 10, wherein the described first non-null set of coefficient is indicate the first sound field the Described first non-null set of the coefficient of the primary vector of one distinct components, wherein one or more described processors are further through matching Set with:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include corresponding in the coefficient to be greater than one or more of more than described second a spherical harmonic coefficients Those of corresponding order of order of basis function coefficient, and exclude corresponding to greater than more than described second in the coefficient At least one coefficient of the order of the order of the basis function corresponding to the one or more in a spherical harmonic coefficient; And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
17. device according to claim 10, wherein the described first non-null set of coefficient is indicate the first sound field the Described first non-null set of the coefficient of the primary vector of one distinct components, wherein one or more described processors are further through matching Set with:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include all coefficients, except at least one of described coefficient;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
18. device according to claim 10, wherein one or more described processors are configured to be based in bit stream with letter Number value indicated and determine the one in the multiple configuration mode.
19. a kind of for handling the device of audio data comprising:
For obtaining the device for indicating the first non-null set of the coefficient of the vector of distinct components of sound field, the vector is from retouching The multiple spherical harmonic coefficients for stating the sound field linearly decompose;
The device of the described first non-null set for packing coefficient;
For obtaining the device of one of multiple configuration modes, so as to being mentioned according to the one in the multiple configuration mode The non-null set of the coefficient of the vector is taken, wherein the one in the multiple configuration mode indicates the coefficient The non-null set includes all coefficients;And
For extracted based on one of described acquisition in the multiple configuration mode the vector the coefficient it is described non- The device of null set.
20. a kind of non-transitory computer-readable storage media has the instruction being stored thereon, described instruction is when being executed Cause one or more processors:
Determine the first non-null set for indicating the coefficient of the vector of distinct components of sound field, the vector is from the description sound field Multiple spherical harmonic coefficients linearly decompose;
One of multiple configuration modes are obtained, so as to extracting the vector according to the one in the multiple configuration mode Coefficient the non-null set, wherein the one in the multiple configuration mode indicates the non-null set of the coefficient Closing includes all coefficients;And
The non-null set of the coefficient of the vector is extracted based on one of the acquisition in the multiple configuration mode It closes.
21. a kind of method for handling audio data comprising:
The the first non-null set for indicating the coefficient of the vector of distinct components of sound field is obtained, the vector is from the description sound field Multiple spherical harmonic coefficients linearly decompose;
One of multiple configuration modes are obtained, so as to extracting the vector according to the one in the multiple configuration mode Coefficient the non-null set, wherein in the multiple configuration mode the one instruction coefficient the non-null set packet Containing all coefficients, except at least one of described coefficient;And
The non-null set of the coefficient of the vector is extracted based on one of the acquisition in the multiple configuration mode It closes.
22. according to the method for claim 21, further comprising the described first non-null set work for extracting the coefficient For the first part of the vector.
23. the method according to claim 11, further comprising:
The described first non-null set of the vector is extracted from side channel information;And
Described first non-null set of the coefficient based on the vector obtain the multiple spherical harmonic coefficient through reassembling into version This.
24. according to the method for claim 21, wherein the vector includes using linear decomposition from the humorous system of the multiple ball The vector that number linearly decomposes.
25. according to the method for claim 24, wherein the linear decomposition includes singular value decomposition.
26. according to the method for claim 21, wherein the described first non-null set of coefficient be the of the first sound field of expression Described first non-null set of the coefficient of the primary vector of one distinct components, the method further includes:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include all coefficients;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
27. according to the method for claim 21, wherein the described first non-null set of coefficient be the of the first sound field of expression Described first non-null set of the coefficient of the primary vector of one distinct components, the method further includes:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include corresponding in the coefficient to be greater than one or more of more than described second a spherical harmonic coefficients Those of corresponding order of order of basis function coefficient;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
28. according to the method for claim 21, wherein the described first non-null set of coefficient be the of the first sound field of expression Described first non-null set of the coefficient of the primary vector of one distinct components, the method further includes:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include corresponding in the coefficient to be greater than one or more of more than described second a spherical harmonic coefficients Those of corresponding order of order of basis function coefficient, and exclude corresponding to greater than more than described second in the coefficient At least one coefficient of the order of the order of the basis function corresponding to the one or more in a spherical harmonic coefficient; And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
29. according to the method for claim 21, wherein obtaining the one in the multiple configuration mode includes being based on The one in the multiple configuration mode is obtained with the value that signal indicates in bit stream.
30. a kind of for handling the device of audio data comprising:
One or more processors, are configured to:
The the first non-null set for indicating the coefficient of the vector of distinct components of sound field is obtained, the vector is from the description sound field Multiple spherical harmonic coefficients linearly decompose;
One of multiple configuration modes are obtained, so as to extracting the vector according to the one in the multiple configuration mode Coefficient the non-null set, wherein in the multiple configuration mode the one instruction coefficient the non-null set packet Containing all coefficients, except at least one of described coefficient;And
The non-null set of the coefficient of the vector is extracted based on one of the acquisition in the multiple configuration mode It closes.
31. device according to claim 30, wherein one or more described processors are further configured to described in extraction First part of the described first non-null set of coefficient as the vector.
32. device according to claim 30, wherein one or more described processors are further configured to lead to from side Described first non-null set of vector described in road information extraction, and the described first non-null set of the coefficient based on the vector Close obtain the multiple spherical harmonic coefficient through reassembling into version.
33. device according to claim 30, wherein the vector includes using linear decomposition from the humorous system of the multiple ball The vector that number linearly decomposes.
34. device according to claim 33, wherein the linear decomposition includes singular value decomposition.
35. device according to claim 30, wherein the described first non-null set of coefficient is indicate the first sound field the Described first non-null set of the coefficient of the primary vector of one distinct components, wherein one or more described processors are further through matching Set with:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include all coefficients;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
36. device according to claim 30, wherein the described first non-null set of coefficient is indicate the first sound field the Described first non-null set of the coefficient of the primary vector of one distinct components, wherein one or more described processors are further through matching Set with:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include corresponding in the coefficient to be greater than one or more of more than described second a spherical harmonic coefficients Those of corresponding order of order of basis function coefficient;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
37. device according to claim 30, wherein the described first non-null set of coefficient is indicate the first sound field the Described first non-null set of the coefficient of the primary vector of one distinct components, wherein one or more described processors are further through matching Set with:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include corresponding in the coefficient to be greater than one or more of more than described second a spherical harmonic coefficients Those of corresponding order of order of basis function coefficient, and exclude corresponding to greater than more than described second in the coefficient At least one coefficient of the order of the order of the basis function corresponding to the one or more in a spherical harmonic coefficient; And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
38. device according to claim 30, wherein one or more described processors are configured to be based in bit stream with letter Number value indicated and determine the one in the multiple configuration mode.
39. a kind of for handling the device of audio data comprising:
For obtaining the device for indicating the first non-null set of the coefficient of the vector of distinct components of sound field, the vector is from retouching The multiple spherical harmonic coefficients for stating the sound field linearly decompose;
The device of the described first non-null set for packing coefficient;
For obtaining the device of one of multiple configuration modes, so as to being mentioned according to the one in the multiple configuration mode The non-null set of the coefficient of the vector is taken, wherein the one instruction coefficient in the multiple configuration mode is described Non- null set includes all coefficients, except at least one of described coefficient;And
For extracted based on one of described acquisition in the multiple configuration mode the vector the coefficient it is described non- The device of null set.
40. a kind of non-transitory computer-readable storage media, has the instruction being stored thereon, described instruction is being performed When cause one or more processors:
Determine the first non-null set for indicating the coefficient of the vector of distinct components of sound field, the vector is from the description sound field Multiple spherical harmonic coefficients linearly decompose;
One of multiple configuration modes are obtained, so as to extracting the vector according to the one in the multiple configuration mode Coefficient the non-null set, wherein in the multiple configuration mode the one instruction coefficient the non-null set packet Containing all coefficients, except at least one of described coefficient;And
The non-null set of the coefficient of the vector is extracted based on one of the acquisition in the multiple configuration mode It closes.
41. a kind of method for handling audio data comprising:
The the first non-null set for indicating the coefficient of the vector of distinct components of sound field is obtained, the vector is from the description sound field Multiple spherical harmonic coefficients linearly decompose;
One of multiple configuration modes are obtained, so as to extracting the vector according to the one in the multiple configuration mode Coefficient the non-null set, wherein the one in the multiple configuration mode indicates the non-null set of the coefficient Close the rank for being greater than basis function corresponding to one or more of the multiple spherical harmonic coefficient comprising corresponding in the coefficient Those of several order coefficient, and exclude corresponding to greater than described one or more in the multiple spherical harmonic coefficient in the coefficient At least one coefficient of the order of the order of the basis function corresponding to person;And
The non-null set of the coefficient of the vector is extracted based on one of the acquisition in the multiple configuration mode It closes.
42. according to the method for claim 41, further comprising the described first non-null set work for extracting the coefficient For the first part of the vector.
43. the method according to claim 11, further comprising:
The described first non-null set of the vector is extracted from side channel information;And
Described first non-null set of the coefficient based on the vector obtain the multiple spherical harmonic coefficient through reassembling into version This.
44. according to the method for claim 41, wherein the vector includes using linear decomposition from the humorous system of the multiple ball The vector that number linearly decomposes.
45. according to the method for claim 44, wherein the linear decomposition includes singular value decomposition.
46. according to the method for claim 41, wherein the described first non-null set of coefficient be the of the first sound field of expression Described first non-null set of the coefficient of the primary vector of one distinct components, the method further includes:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include all coefficients;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
47. according to the method for claim 41, wherein the described first non-null set of coefficient be the of the first sound field of expression Described first non-null set of the coefficient of the primary vector of one distinct components, the method further includes:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include corresponding in the coefficient to be greater than one or more of more than described second a spherical harmonic coefficients Those of corresponding order of order of basis function coefficient;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
48. according to the method for claim 41, wherein the described first non-null set of coefficient be the of the first sound field of expression Described first non-null set of the coefficient of the primary vector of one distinct components, the method further includes:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include all coefficients, except at least one of described coefficient;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
49. according to the method for claim 41, wherein obtaining the one in the multiple configuration mode includes being based on The one in the multiple configuration mode is obtained with the value that signal indicates in bit stream.
50. a kind of for handling the device of audio data comprising:
One or more processors, are configured to:
The the first non-null set for indicating the coefficient of the vector of distinct components of sound field is obtained, the vector is from the description sound field Multiple spherical harmonic coefficients linearly decompose;
One of multiple configuration modes are obtained, so as to extracting the vector according to the one in the multiple configuration mode Coefficient the non-null set, wherein the one in the multiple configuration mode indicates the non-null set of the coefficient Close the rank for being greater than basis function corresponding to one or more of the multiple spherical harmonic coefficient comprising corresponding in the coefficient Those of several order coefficient, and exclude corresponding to greater than described one or more in the multiple spherical harmonic coefficient in the coefficient At least one coefficient of the order of the order of the basis function corresponding to person;And
The non-null set of the coefficient of the vector is extracted based on one of the acquisition in the multiple configuration mode It closes.
51. device according to claim 50, wherein one or more described processors are further configured to described in extraction First part of the described first non-null set of coefficient as the vector.
52. device according to claim 50, wherein one or more described processors are further configured to lead to from side Described first non-null set of vector described in road information extraction, and the described first non-null set of the coefficient based on the vector Close obtain the multiple spherical harmonic coefficient through reassembling into version.
53. device according to claim 50, wherein the vector includes using linear decomposition from the humorous system of the multiple ball The vector that number linearly decomposes.
54. device according to claim 53, wherein the linear decomposition includes singular value decomposition.
55. device according to claim 50, wherein the described first non-null set of coefficient is indicate the first sound field the Described first non-null set of the coefficient of the primary vector of one distinct components, wherein one or more described processors are further through matching Set with:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include all coefficients;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
56. device according to claim 50, wherein the described first non-null set of coefficient is indicate the first sound field the Described first non-null set of the coefficient of the primary vector of one distinct components, wherein one or more described processors are further through matching Set with:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include corresponding in the coefficient to be greater than one or more of more than described second a spherical harmonic coefficients Those of corresponding order of order of basis function coefficient;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
57. device according to claim 50, wherein the described first non-null set of coefficient is indicate the first sound field the Described first non-null set of the coefficient of the primary vector of one distinct components, wherein one or more described processors are further through matching Set with:
Obtain the second non-null set for indicating the coefficient of the secondary vector of the second distinct components of the second sound field, the secondary vector It is linearly decomposed from more than the second a spherical harmonic coefficients for describing second sound field;
One of a configuration mode more than second is obtained, so as to extracting according to the one in more than described second a configuration modes Described second non-null set of the coefficient of the vector, wherein the one more than described second in a configuration mode indicates coefficient The described second non-null set include all coefficients, except at least one of described coefficient;And
The institute of the coefficient of the secondary vector is extracted based on one of the acquisition in more than described second a configuration modes State the second non-null set.
58. device according to claim 50, wherein one or more described processors are configured to be based in bit stream with letter Number value indicated and determine the one in the multiple configuration mode.
59. a kind of for handling the device of audio data comprising:
For obtaining the device for indicating the first non-null set of the coefficient of the vector of distinct components of sound field, the vector is from retouching The multiple spherical harmonic coefficients for stating the sound field linearly decompose;
The device of the described first non-null set for packing coefficient;
For obtaining the device of one of multiple configuration modes, so as to being mentioned according to the one in the multiple configuration mode The non-null set of the coefficient of the vector is taken, wherein the one in the multiple configuration mode indicates the coefficient The non-null set includes corresponding in the coefficient to be greater than base corresponding to one or more of the multiple spherical harmonic coefficient Those of the order of order of bottom function coefficient, and exclude corresponding to greater than in the multiple spherical harmonic coefficient in the coefficient At least one coefficient of the order of the order of the basis function corresponding to the one or more;And
For extracted based on one of described acquisition in the multiple configuration mode the vector the coefficient it is described non- The device of null set.
60. a kind of non-transitory computer-readable storage media, has the instruction being stored thereon, described instruction is being performed When cause one or more processors:
Determine the first non-null set for indicating the coefficient of the vector of distinct components of sound field, the vector is from the description sound field Multiple spherical harmonic coefficients linearly decompose;
One of multiple configuration modes are obtained, so as to extracting the vector according to the one in the multiple configuration mode Coefficient the non-null set, wherein the one in the multiple configuration mode indicates the non-null set of the coefficient Close the rank for being greater than basis function corresponding to one or more of the multiple spherical harmonic coefficient comprising corresponding in the coefficient Those of several order coefficient, and exclude corresponding to greater than described one or more in the multiple spherical harmonic coefficient in the coefficient At least one coefficient of the order of the order of the basis function corresponding to person;And
The non-null set of the coefficient of the vector is extracted based on one of the acquisition in the multiple configuration mode It closes.
CN201480031162.XA 2013-05-29 2014-05-29 The compression through exploded representation of sound field CN105340008B (en)

Priority Applications (39)

Application Number Priority Date Filing Date Title
US201361828445P true 2013-05-29 2013-05-29
US201361828615P true 2013-05-29 2013-05-29
US61/828,615 2013-05-29
US61/828,445 2013-05-29
US201361829155P true 2013-05-30 2013-05-30
US201361829174P true 2013-05-30 2013-05-30
US201361829182P true 2013-05-30 2013-05-30
US61/829,155 2013-05-30
US61/829,174 2013-05-30
US61/829,182 2013-05-30
US201361829791P true 2013-05-31 2013-05-31
US201361829846P true 2013-05-31 2013-05-31
US61/829,791 2013-05-31
US61/829,846 2013-05-31
US201361886605P true 2013-10-03 2013-10-03
US201361886617P true 2013-10-03 2013-10-03
US61/886,617 2013-10-03
US61/886,605 2013-10-03
US201361899034P true 2013-11-01 2013-11-01
US201361889041P true 2013-11-01 2013-11-01
US61/899,034 2013-11-01
US61/889,041 2013-11-01
US201461925126P true 2014-01-08 2014-01-08
US201461925112P true 2014-01-08 2014-01-08
US201461925158P true 2014-01-08 2014-01-08
US201461925074P true 2014-01-08 2014-01-08
US61/925,112 2014-01-08
US61/925,158