KR20090037806A - Encoding and decoding method using variable subband aanlysis and apparatus thereof - Google Patents

Abstract

An encoding and decoding method using variable subband analysis and an apparatus thereof are provided to increase the number of subbands while minimizing the increase of bit races, thereby outputting more superior sound. Audio objects(1~M) are inputted to an audio encoding unit and a frequency conversion unit. The audio encoding unit down-mixes the audio object. The frequency conversion unit(202) converts an audio object into a frequency domain. A subband configuration unit(203) subdivides a subband of a frequency-converted signal to variable subbands. A parameter generation unit(205) extracts parameters necessary for restoring the audio objects from the down mix signal. An encoding unit(206) encodes parameter information including a parameter generated from a parameter generating unit.

Description

Encoding and decoding method using variable subband analysis and apparatus therefor {Encoding and Decoding Method using Variable Subband Aanlysis and Apparatus}

The present invention relates to an encoding and decoding method and apparatus, and more particularly, to an encoding and decoding method and apparatus using variable subband analysis.

The present invention is derived from a study performed as part of the support for the development of information and communication standards of the Ministry of Information and Communication and the Institute for Information and Telecommunication Research Promotion. [Task Management Number: 2007-S-004-01, Title: Development of Glassless Single-User 3D Broadcasting Technologies]

According to the prior art, a plurality of audio objects composed of various channels cannot be variously combined according to a user's needs, and thus one audio content cannot be consumed in various forms. As a result, the user can only consume audio content passively.

MPEG audio subgroups, one of the standardization groups, have been developing audio coding standards such as AAC (Advanced Audio Coding) and MPEG Surround (MPS). AAC is a technique for high quality audio encoding for mono or stereo channel signals, and MPS is a technique suitable for multi-channel audio encoding. As with these techniques, conventional audio coders have been focused on channel based audio signals. For example, Spatial Audio Coding (SAC), which is a recently introduced spatial cue based audio encoding method, is an example.

According to the conventional spatial audio coding (SAC) technology, a multichannel audio signal is encoded into a downmixed mono channel or stereo channel signal and spatial cue information, and a high quality multichannel signal is transmitted even at a low bit rate. According to the SAC technology, an audio signal is analyzed for each subband, and an original multichannel audio signal is recovered from the downmixed mono channel or stereo channel signal based on spatial cue information corresponding to each subband. The spatial cue information includes information for reconstruction of the original signal in the decoding process, and determines the sound quality of the audio signal reproduced in the SAC decoding apparatus. MPEG is a standardization of SAC technology under the name of MPEG Surround (MPS), and uses CLD (Channel Level Difference) as a spatial cue.

According to the SAC, as a multichannel audio signal, only one audio object can be encoded and decoded, so that a multi-object audio signal composed of multiple channels, for example, audio of various objects composed of mono channels, stereo channels, and 5.1 channels The signal cannot be encoded and decoded.

According to another conventional Binaural Cue Coding (BCC) technique, since a multi-object audio signal composed of only a mono channel can be encoded and decoded, a multi-object audio signal composed of multiple channels other than a mono channel is generated. It cannot be encoded and decoded.

Therefore, when using a conventional audio encoding scheme, if the transmission of various audio objects including mono, stereo or multi-channel signals is required, a problem of high bitrate is inevitable.

In order to solve this problem, another audio coding scheme called Spatial Audio Object Coding (SAOC) has been proposed. If the conventional SAC scheme focuses on multichannel audio encoding, the SAOC scheme is a method for encoding multi-object audio. Multi-object audio encoding is a technique for compressing and transmitting different audio objects. In this case, a spatial cue representative of characteristics of each object may be extracted. Thus, while conventional audio coding schemes are techniques for compressing each object separately, SAOC is a technique for simultaneously processing one or more multiple objects. In SAOC, one or more audio objects are represented by one stereo downmixed signal and side information. Therefore, when using SAOC, the bit rate can be significantly reduced as compared with the conventional audio encoder.

In the conventional audio service, there was a general functional restriction that the user has to listen passively to the transmitted audio content. However, the audio coding method for each object is to provide a more active service to the user. Accordingly, not only can control each audio object, but also various audio services and contents can be created from one content combination. SAOC, on the other hand, can apply a mixer / renderer that provides functions such as panning, attenuation, and suppression, thereby allowing interaction with the user. Accordingly, the audio object can be flexibly controlled.

However, the SAC-based SAOC coding scheme cannot completely restore the original sound source, and therefore, the control performance of key functions such as Karaoke and Solo-representation is limited. In particular, such a SAOC system has a problem of causing severe degradation of sound quality due to a downmix process based on a limited number of subbands. Therefore, to solve this problem, the number of subbands must be increased. However, when the number of subbands is indiscriminately increased, additional information to be transmitted also increases.

Accordingly, an object of the present invention is to provide an encoding and decoding method and apparatus capable of improving sound quality by subdividing a subband structure while minimizing an increase in bitlace.

Other objects and advantages of the present invention can be understood by the following description, and will be more clearly understood by the embodiments of the present invention. It will also be appreciated that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the claims.

The encoding method using the variable subband analysis according to an embodiment of the present invention for solving the above-mentioned problems is a step of converting the audio object to the frequency domain, subband according to the characteristics of the subband of the signal converted to the frequency domain Subdividing into variable subbands, generating variable subband information including information about the subdivided variable subbands, and generating parameter information used for reconstruction of an audio object based on the variable subbands.

According to another embodiment of the present invention, an encoding apparatus using variable subband analysis includes a frequency converter for converting an audio object into a frequency domain, and subdividing the subband into variable subbands according to characteristics of a subband of a signal converted into a frequency domain. And a subband constructing unit for generating variable subband information including information about the subdivided variable subbands, and a parameter generating unit for generating parameter information used for reconstruction of an audio object based on the variable subbands.

According to another embodiment of the present invention, a decoding method using variable subband analysis includes variable subband information including variable subband information divided into subbands according to characteristics of subbands of an audio object, and variable subbands. Receiving a bitstream including parameter information for reconstructing the audio object based on the data, reconstructing the subband based on the variable subband information, and reconstructing the audio object using the parameter information.

According to another embodiment of the present invention, a decoding apparatus using variable subband analysis may include variable subband information and variable subband information including information about a variable subband subdivided into subbands according to characteristics of a subband of an audio object. A receiver for receiving a bitstream including parameter information for reconstructing the audio object based on the data, a subband reconstruction unit for reconstructing the subband based on variable subband information, and a reconstruction unit for reconstructing the audio object using parameter information Include.

The encoding method using variable subband analysis according to another embodiment of the present invention comprises the steps of generating and encoding a plurality of input audio objects as a downmix signal, converting the plurality of audio objects to a frequency domain, and a frequency domain. Subdividing the subbands of the converted signal into variable subbands according to characteristics of the subbands, and generating variable subband information including information about the subdivided variable subbands; Generating parameter information used for reconstruction and encoding variable subband information and parameter information.

An encoding apparatus using variable subband analysis according to another embodiment of the present invention includes an audio encoder for generating and encoding a plurality of input audio objects as a downmix signal, and a frequency converter for converting the plurality of audio objects to a frequency domain. A subband configuration unit for subdividing a subband of a signal converted into a frequency domain into a variable subband according to characteristics of a subband and generating variable subband information including information about the subdivided variable subband, and a variable subband A parameter generator for generating parameter information used for reconstruction of the downmix signal, and an encoder for encoding variable subband information and parameter information.

According to another embodiment of the present invention, a decoding method using variable subband analysis may include a downmix signal for a plurality of audio objects from an input bitstream and a variable subband subdivided according to characteristics of subbands of the plurality of audio objects. Decoding the variable subband information including information about the information and parameter information for reconstructing the downmix signal based on the variable subband, converting the decoded downmix signal into a frequency domain, and decoded variable subband information Reconstructing the subbands based on the received information, restoring the plurality of audio objects using the decoded parameter information, the downmix signal converted into the frequency domain, and the reconstructed subband, and converting the plurality of restored audio objects into the time domain. Converting.

According to another embodiment of the present invention, a decoding apparatus using variable subband analysis may include a downmix signal for a plurality of audio objects and a variable subband subdivided according to characteristics of subbands of a plurality of audio objects from an input bitstream. A decoder which decodes the variable subband information including the information and parameter information for reconstructing the downmix signal based on the variable subband, a frequency converter which converts the decoded downmix signal into a frequency domain, and a decoded variable subband A subband reconstruction unit for reconstructing the subbands based on the band information, a reconstruction unit for reconstructing a plurality of audio objects using the decoded parameter information, the downmix signal converted to the frequency domain, and the reconstructed subband, and a plurality of reconstructed plurality It includes a time conversion unit for converting the audio object to the time domain.

The above objects, features and advantages will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, whereby those skilled in the art may easily implement the technical idea of the present invention. There will be. In addition, in describing the present invention, when it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

According to the present invention, the sound quality can be improved by subdividing the subband structure of the audio object.

The following merely illustrates the principles of the invention. Therefore, those skilled in the art, although not explicitly described or illustrated herein, can embody the principles of the present invention and invent various devices that fall within the spirit and scope of the present invention. In addition, all conditional terms and embodiments listed herein are in principle clearly intended to be understood solely for the purpose of understanding the concept of the invention and are not to be limited to the specifically listed embodiments and states. do.

In addition, it is to be understood that all detailed descriptions, including the principles, aspects, and embodiments of the present invention, as well as listing specific embodiments, are intended to include structural and functional equivalents of these matters. In addition, these equivalents should be understood to include not only equivalents currently known but also equivalents to be developed in the future, that is, all devices invented to perform the same function regardless of structure.

Thus, for example, it should be understood that the block diagrams herein represent a conceptual view of example circuitry embodying the principles of the invention. Similarly, all flowcharts, state transitions, pseudocodes, and the like are understood to represent various processes performed by a computer or processor, whether or not the computer or processor is substantially illustrated on a computer readable medium and whether the computer or processor is clearly shown. Should be.

The functionality of the various elements shown in the figures, including functional blocks represented by a processor or similar concept, can be provided by the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functionality may be provided by a single dedicated processor, by a single shared processor or by a plurality of individual processors, some of which may be shared.

In addition, the explicit use of terms presented in terms of processor, control, or similar concept should not be interpreted exclusively as a citation to hardware capable of running software, and without limitation, ROM for storing digital signal processor (DSP) hardware, software. (ROM), RAM, and non-volatile memory are to be understood to implicitly include. Other hardware for the governor may also be included.

In the claims of this specification, components expressed as means for performing the functions described in the detailed description include all types of software including, for example, a combination of circuit elements or firmware / microcode, etc. that perform the functions. It is intended to include all methods of performing a function which are combined with appropriate circuitry for executing the software to perform the function. The invention, as defined by these claims, is equivalent to what is understood from this specification, as any means capable of providing such functionality, as the functionality provided by the various enumerated means are combined, and in any manner required by the claims. It should be understood that.

The above objects, features and advantages will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, whereby those skilled in the art may easily implement the technical idea of the present invention. There will be. In addition, in describing the present invention, when it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

In the encoding using the variable subband analysis according to the present invention, by subdividing the subbands of the audio object converted into the frequency domain according to the characteristics of the subbands, it is possible to achieve high quality sound while minimizing an increase in bitrate.

Specifically, according to the encoding using the variable subband analysis according to the present invention, the audio object is transformed into a frequency domain (T / F transform), and the subband is changed into a variable subband according to the characteristics of the subband of the signal transformed into the frequency domain. And subband information including subband subdivided subband information is generated. Generates parameter information used for reconstruction of the audio object based on the variable subband. By subdividing the subbands, a high quality sound quality can be provided, and by subdividing the subbands selectively, it is possible to suppress an increase in the bit rate.

Variable subband information and parameter information are encoded. In this case, the encoding may be performed by lossless coding. The audio object is generated from the downmixed signal and audio encoded. Audio encoding may use a conventional encoding method. The encoded variable subband information, parameter information, and audio object are generated in a bitstream.

The characteristics of the subbands may include dispersion coefficient characteristics of power ratios for each subband. For example, when the dispersion coefficient of the power ratio of a specific subband is greater than or equal to a predetermined threshold, the subband is subdivided. If it is less than a predetermined threshold, the subband is not subdivided and the existing subband is maintained.

Meanwhile, the parameter information may be quantized using a variable bit level, and the parameter information may include spatial parameter information. By quantizing using a variable bit level, it is possible to minimize the bit rate that increases as the number of subbands increases.

The decoding using the variable subband analysis according to the present invention restores the audio object using the information subdivided into subbands of the audio object transformed into the frequency domain according to the characteristics of the subbands, thereby minimizing the increase of the bit rate, Sound quality of can be achieved.

Specifically, according to the decoding using the variable subband analysis according to the present invention, the audio object is based on the variable subband information and the variable subband including information on the variable subband subdivided according to the characteristics of the subband of the audio object. Receives a bitstream including parameter information for reconstruction, reconstructs a subband based on variable subband information, and reconstructs an audio object using parameter information. By subdividing the subbands, a high quality sound quality can be provided, and by subdividing the subbands selectively, it is possible to suppress an increase in the bit rate.

Variable subband information and parameter information are decoded. In this case, the decoding may be performed by lossless decoding. The bitstream may include a bitstream for an audio object, and the audio object is audio decoded. Audio decoding may use a conventional decoding method. The decoded audio object is frequency transformed (T / F transform). The audio object is reconstructed using the frequency-converted audio object and the reconstructed subband. The reconstructed audio object is output after being subjected to F / T transform.

The characteristics of the subbands may include dispersion coefficient characteristics of power ratios for each subband. For example, when the dispersion coefficient of the power ratio of a specific subband is greater than or equal to a predetermined threshold, the subband is subdivided. If it is less than the predetermined threshold, the subband is not subdivided and is maintained as an existing subband. The variable subband information may include the characteristic information of the subband, and the subband is reconfigured using this.

Meanwhile, the parameter information may be inversely quantized using a variable bit level, and the parameter information may include spatial parameter information. By quantizing using a variable bit level, it is possible to minimize the bit rate that increases as the number of subbands increases.

It will be described in detail with the following examples.

<Coding>

The encoding method using the variable subband analysis according to the present invention includes generating and encoding a plurality of input audio objects as a downmix signal, converting the plurality of audio objects into a frequency domain, and sub-processing a signal converted into a frequency domain. Subdividing the band into variable subbands according to characteristics of the subbands, generating variable subband information including information about the subdivided variable subbands, and a parameter used for reconstructing a downmix signal based on the variable subbands Generating information and encoding variable subband information and parameter information. Here, the characteristics of the subband may include a dispersion coefficient characteristic of the power ratio with respect to the subband. Here, the parameter information may include spatial parameter information.

Meanwhile, generating the parameter information may include quantizing the parameter information using a variable bit level.

The encoding apparatus using the variable subband analysis according to the present invention includes an audio encoder for generating and encoding a plurality of input audio objects as a downmix signal, a frequency converter for converting the plurality of audio objects to a frequency domain, and a frequency domain. A subband configuration unit for subdividing the subbands of the divided signals into variable subbands according to characteristics of the subbands, and generating variable subband information including information about the subdivided variable subbands, the downmix based on the variable subbands And a parameter generator for generating parameter information used for reconstruction of the signal, and an encoder for encoding variable subband information and parameter information. Here, the characteristics of the subband may include a dispersion coefficient characteristic of the power ratio with respect to the subband. Here, the parameter information may include spatial parameter information.

The parameter generator may include a quantizer for quantizing parameter information using a variable bit level.

<Decryption>

A decoding method using variable subband analysis according to the present invention includes downmix signals for a plurality of audio objects from input bitstreams, and information about variable subbands divided according to characteristics of subbands of the plurality of audio objects. Decoding parameter information for restoring the downmix signal based on the variable subband information and the variable subband, converting the decoded downmix signal into a frequency domain, and subband based on the decoded variable subband information Reconstructing a plurality of audio objects using the decoded parameter information, the downmix signal converted into the frequency domain, and the reconstructed subband, and converting the restored plurality of audio objects into a time domain. do. Here, the characteristics of the subband may include a dispersion coefficient characteristic of the power ratio with respect to the subband. Here, the parameter information may include spatial parameter information.

Meanwhile, the decoding may include dequantizing parameter information using a variable bit level.

The decoding apparatus using the variable subband analysis according to the present invention includes a downmix signal for a plurality of audio objects from the input bitstream and information about the variable subbands divided according to the characteristics of the subbands of the plurality of audio objects. A decoder which decodes parameter information for reconstructing the downmix signal based on the variable subband information and the variable subband, a frequency converter which converts the decoded downmix signal into a frequency domain, and based on the decoded variable subband information. A subband reconstruction unit for reconstructing the subbands, a decoded parameter information, a downmix signal converted into a frequency domain, and a reconstruction unit for reconstructing a plurality of audio objects using the reconstructed subbands and a plurality of restored audio objects in a time domain It includes a time conversion unit to convert to. Here, the characteristics of the subband may include a dispersion coefficient characteristic of the power ratio with respect to the subband. Here, the parameter information may include spatial parameter information.

The decoder may include an inverse quantizer that inversely quantizes the parameter information using the variable bit level.

Hereinafter, with reference to the drawings, a specific embodiment will be described.

In the general multi-object / multi-channel audio encoding process, since the downmix process is performed as one object, the restoration of each object cannot be completely performed during the decoding process. In particular, when the power of one object signal is completely reduced, such as in a karaoke mode, the sound quality deterioration is noticeably worse.

Accordingly, the present invention extracts more accurate parameters by variably increasing the number of subbands analyzing parameters in the multi-object / multi-channel audio encoding / decoding process, and consequently recovers objects more clearly from the downmix signal. In this process, it is possible to minimize the increase in the bit rate by applying different quantization levels according to the frequency characteristics of the signals.

1 illustrates an audio encoding / decoding process according to the present invention. The encoder 101 receives an audio object. The input audio object is not limited. Therefore, a plurality of audio objects (Object # 1, Object # 2, Object # 3 ...) may be input. The encoder 101 generates a downmix signal using the input objects, and extracts parameters required in the decoding process. Side information of FIG. 1 may be included in a parameter. The decoder 102 performs decoding and outputs an audio object reconstructed from the encoder 101 using the downmix signal and the parameters. The reconstructed audio object is controlled by the mixer / renderer (Mixer / Renderer 103) such as position / level and the like (Chan. # 1, Chan. # 2, Chan. # 3, ...). Is output. Here, the encoder 101 and the decoder 102 may use the SAOC method.

2 is a block diagram of a multi-object audio encoder according to the present invention. Multi-object audio encoding according to the present invention analyzes the frequency band characteristics of signals, defines a subband structure for analyzing parameters, and applies a quantization method of parameters differently according to frequency characteristics. The defined subband structure is reconstructed for reconstruction in the decoding process.

The audio objects 1, 2,..., M are input to the audio encoder 201 and the frequency transform unit 202. The audio code lower portion 201 downmixes and encodes the audio objects 1, 2, ..., M. The frequency converter 202 converts the audio objects 1, 2, ..., M into a frequency domain.

The subband configuration unit 203 subdivides the subbands of the frequency-converted signal into variable subbands according to the characteristics of the subbands. The parameter generator 205 extracts parameters necessary for audio objects to be recovered from the downmix signal in the decoding process based on the variable subband.

The parameter for the subband may include IOLD (Inter-Object Level Difference) information. IOLD is a parameter that calculates the ratio of power between two objects for each subband. The expression of IOLD expressing this is shown in [Equation 1].

[Equation 1]

Here, M is the number of subbands, and k and b are frequency coefficients and subband indexes, respectively. In addition, the molecular terms and denominator terms can be defined interchangeably. The subband may have a fixed subband according to an encoding method. For example, in MPEG Surround, 20 or 28 fixed subbands are applied to one audio signal frame. In the case of calculating for each fixed subband, if the resolution of the band analyzing this is low, there is a disadvantage in that the two signals cannot be separated well in the decoding process. In the present invention, in order to improve this, the subband configuration unit 203 configures the subband into more subdivided variable subbands, so that more accurate parameters can be analyzed. The subband configuration unit 203 will be described in more detail with reference to FIG. 4.

The encoder 204 encodes the variable subband information generated by the subband configuration unit 203. The encoder 206 encodes parameter information including the parameter generated by the parameter generator. The encoders 204 and 206 may use a lossless coding scheme. In the bitstream formatter 207, encoded variable subband information, parameter information, and an audio object are generated as a bitstream. Here, the bitstream may be a SAOC bitstream.

4 is a view for explaining the configuration of a variable subband according to the present invention. The subband configuration unit 203 of FIG. 2 may include a spectrum analysis unit 401 of FIG. 4. In FIG. 4, an object to be freely controlled by the user is referred to as Object # 1, and other objects are referred to as Object # 2, Object # 3,. Here, the spectrum analyzer 401 analyzes the power of the frequency band of each signal and outputs variable subband information which is new subband information.

The basic structure of the subbands to be analyzed for parameters follows the 28 bands used in MPEG Surround. When the power ratio of the two signals in each subband changes significantly, the specific band is further subdivided. This condition is expressed by the equation as shown in [Equation 2] and [Equation 3].

[Equation 2]

[Equation 3]

Here, avrgb is a variable representing an average of power ratios of two signals in a b-th subband, and varb is a dispersion coefficient representing a degree of change of power ratios of two signals.

When the dispersion coefficient according to [Equation 3] exceeds a predetermined threshold, the analysis band (b-th subband) is further subdivided to form an overall subband, and additionally transmits a parameter indicating the structure of the variable subband thus configured. In the decoding step, it is possible to easily configure the granular variable subband. For example, a parameter indicating the structure of a subband is represented by 0 or 1 for each band. A subband labeled with 1 means that the band needs to be further subdivided.

3 is a block diagram of a multi-object audio decoder according to the present invention. The bitstream demultiplexer 301 receives a bitstream, separates a signal for an audio object, a signal for parameter information, and a signal for variable subband information, respectively, and decodes each of the decoders 302, 304, and 305. Will output Here, the bitstream may be a SAOC bitstream. The signal for the audio object is decoded by the audio decoder 302 and output as a downmixed signal. The downmixed signal is frequency converted by the frequency converter 303. The signal for the parameter information is decoded by the parameter decoder 304 and output to the decompression unit 307. The signal for the variable subband information is decoded by the variable subband decoder 305 and output to the subband reconstructor 306. The decoding of the signal for the parameter information and the signal for the variable subband information may use a lossless decoding method.

The reconstruction unit 307 reconstructs the audio object by using the downmix signal and parameter information of the frequency-converted audio object based on the subband reconstructed using the variable subband information in the subband reconstruction unit 306. In this case, the parameter information may be a spatial parameter including spatial queue information. The restored audio object is converted into a time domain by the time converter and finally output as an audio object. For example, restoring two objects from one downmix signal using the IOLD parameter may use Equation 4 below.

[Equation 4]

5 is a diagram for explaining reconfiguration of a subband according to the present invention. 5 illustrates a process of reconstructing the shape of a subband used in an encoding process by the subband reconstruction unit 306 of FIG. 3. 28 subbands are represented by 0 or 1, respectively, according to the variable subband information. In the case of the band indicated by 0, the existing one is used as it is, and in the case of the band indicated by 1, the band is regularly divided by a predetermined number. Block 501 indicates partition information of 28 subbands used in MPEG Surround on the basis of FFT, and A (k) as the output indicates the partition of the k-th band.

6 is a diagram for explaining quantization using a variable bit level according to the present invention. The variable level quantization unit 601 may be included in the parameter generator 205 of FIG. 2. The variable level quantization unit 601 analyzes frequency band characteristics of the input parameter and performs variable bit quantization according to the input parameter to output the quantized parameter. By performing variable bit quantization in the parameter generator, an increase in bit rate can be minimized.

7 is a view for explaining inverse quantization using a variable bit level according to the present invention. The variable level dequantizer 701 may be included in the variable subband decoder 305 of FIG. 3. The variable level inverse quantization unit 701 receives the quantized parameter and performs variable bit inverse quantization according to the frequency band characteristic to output the dequantized parameter.

As described above, the method of the present invention may be implemented as a program and stored in a recording medium (CD-ROM, RAM, ROM, floppy disk, hard disk, magneto-optical disk, etc.) in a computer-readable form. Since this process can be easily implemented by those skilled in the art will not be described in more detail.

The present invention described above is capable of various substitutions, modifications, and changes without departing from the technical spirit of the present invention for those skilled in the art to which the present invention pertains. It is not limited by the drawings.

The present invention is used for encoding and decoding audio.

1 illustrates an audio encoding / decoding process according to the present invention.

2 is a block diagram of a multi-object audio encoder according to the present invention.

3 is a block diagram of a multi-object audio encoder according to the present invention.

4 is a view for explaining the configuration of a variable subband according to the present invention.

5 is a diagram for explaining reconfiguration of a subband according to the present invention.

6 is a diagram for explaining quantization using a variable bit level according to the present invention.

7 is a view for explaining inverse quantization using a variable bit level according to the present invention.

Claims (20)

Generating and encoding a plurality of input audio objects as a downmix signal; Converting the plurality of audio objects into a frequency domain; Subdividing a subband of the signal converted into the frequency domain into a variable subband according to characteristics of the subband, and generating variable subband information including information about the subdivided variable subband; Generating parameter information used for reconstruction of the downmix signal based on the variable subband; And And encoding the variable subband information and the parameter information. The method of claim 1, The characteristics of the subband And a coefficient of variance characteristic of a power ratio for the subband. The method of claim 1, Generating the parameter information And quantizing the parameter information using a variable bit level. The method of claim 1, The parameter information is An encoding method using variable subband analysis, including spatial parameter information. An audio encoder for generating and encoding a plurality of input audio objects as a downmix signal; A frequency converter converting the plurality of audio objects into a frequency domain; A subband configuration unit for subdividing the subbands of the signal converted into the frequency domain into variable subbands according to characteristics of the subbands, and generating variable subband information including information about the subdivided variable subbands; A parameter generator for generating parameter information used to recover the downmix signal based on the variable subband; And And an encoding unit encoding the variable subband information and the parameter information. The method of claim 5, The characteristics of the subband And a coefficient of variance characteristic of the power ratio for the subband. The method of claim 5, The parameter generation unit And a quantizer for quantizing the parameter information using a variable bit level. The method of claim 5, The parameter information is An encoding apparatus using variable subband analysis including spatial parameter information. Based on the downmix signal for the plurality of audio objects from the input bitstream, variable subband information including information on the variable subbands subdivided according to the characteristics of the subbands of the plurality of audio objects, and the variable subband. Decoding parameter information for recovering the downmix signal; Converting the decoded downmix signal into a frequency domain; Reconstructing a subband based on the decoded variable subband information; Restoring a plurality of audio objects using the decoded parameter information, the downmix signal converted into the frequency domain, and the reconstructed subband; And And converting the reconstructed plurality of audio objects into a time domain. The method of claim 9, The characteristics of the subband And a dispersion coefficient characteristic of power ratio for the subband. The method of claim 9, The decoding step And inverse quantizing the parameter information using a variable bit level. The method of claim 9, The parameter information is A decoding method using variable subband analysis including spatial parameter information. Based on the downmix signal for the plurality of audio objects from the input bitstream, variable subband information including information on the variable subbands subdivided according to the characteristics of the subbands of the plurality of audio objects, and the variable subband. A decoder which decodes parameter information for restoring the downmix signal; A frequency converter converting the decoded downmix signal into a frequency domain; A subband reconstruction unit reconstructing a subband based on the decoded variable subband information; A reconstruction unit for reconstructing a plurality of audio objects using the decoded parameter information, the downmix signal converted into the frequency domain, and the reconstructed subband; And And a time converter for converting the restored plurality of audio objects into a time domain. The method of claim 13, The characteristics of the subband And a dispersion coefficient characteristic of a power ratio for the subband. The method of claim 13, The decoding unit And an inverse quantization unit for inversely quantizing the parameter information by using a variable bit level. The method of claim 13, The parameter information is A decoding apparatus using variable subband analysis including spatial parameter information. Converting the audio object into a frequency domain; Subdividing the subband into variable subbands according to characteristics of subbands of the signal converted into the frequency domain, and generating variable subband information including information about the subdivided variable subbands; And Generating parameter information used for reconstruction of the audio object based on the variable subband. A frequency converter converting the audio object into a frequency domain; A subband configuration unit configured to subdivide the subband into a variable subband according to characteristics of a subband of the signal converted into the frequency domain, and generate variable subband information including information about the subdivided variable subband; And And a parameter generator for generating parameter information used for reconstruction of the audio object based on the variable subband. A bitstream including variable subband information including information on a variable subband subdivided in the subband according to characteristics of a subband of an audio object and parameter information for reconstructing the audio object based on the variable subband. Receiving; Reconstructing subbands based on the variable subband information; And And reconstructing the audio object using the parameter information. A bitstream including variable subband information including information on a variable subband subdivided in the subband according to characteristics of a subband of an audio object and parameter information for reconstructing the audio object based on the variable subband. Receiving unit for receiving; A subband reconstruction unit for reconstructing a subband based on the variable subband information; And And a reconstruction unit for reconstructing the audio object by using the parameter information.

Images