CN104347077A - Stereo coding method and stereo decoding method - Google Patents
Stereo coding method and stereo decoding method Download PDFInfo
- Publication number
- CN104347077A CN104347077A CN201410573759.9A CN201410573759A CN104347077A CN 104347077 A CN104347077 A CN 104347077A CN 201410573759 A CN201410573759 A CN 201410573759A CN 104347077 A CN104347077 A CN 104347077A
- Authority
- CN
- China
- Prior art keywords
- block
- stereo parameter
- stereo
- monophony
- coded data
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
Landscapes
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention discloses a stereo coding method which comprises the following steps: acquiring a time-domain signal, selecting a block length, converting time frequency, coding stereo parameters, processing a single track and packaging. The invention also discloses a stereo decoding method which comprises the following steps: unpackaging, decoding the single track, decoding the stereo parameters, processing stereo and inversely converting the time frequency. After the stereo coding method and the stereo decoding method are adopted, an integration framework which is more compact with a core single track coding and encoding device is provided, and the step of converting the time frequency is reused, so that the calculated amount and the time delay cost which are caused by repeatedly converting the time frequency can be reduced; furthermore, a more flexible coding method of interframe combination and intra-frame combination is provided, so that the data size of the final coding result is reduced, and the higher audio quality can be obtained under the different stereo parameter code rate configurations.
Description
Technical field
The present invention relates to audio-visual technique field, relate to a kind of stereo coding/decoding method specifically.
Background technology
In the prior art, owing to there is stronger advantage with numerical approach recording voice ratio analogy method recording voice, such as during transmission noise resisting ability strong, increase audio dynamic range, repeatedly pirate recordings and there is no signal attenuation etc.Therefore numerical approach record audio file is usually adopted.But along with the requirement of people to multimedia audio is more and more higher, the data volume needed for recording voice is increasing.The audio file of big data quantity not only needs to take very wide transmission bandwidth in transmitting procedure, and needs to take a large amount of storage spaces in storing process.For solving the excessive problem of audio file data amount, usually in DAB, adopting Digital Audio Compression Technology, voice data is compressed.
Usually many audio coding methods based on frequency domain are by adopting the time-frequency conversion module of adaptive block length, provide sound signal optimal time frequency resolution, obtain higher compression coding gain.Such as, at advanced audio (Advanced Audio Coding, AAC) adopt two kinds of filter bank structure of the long block of 1024 and the short block combination of 128 in, respectively stationary signal and momentary signal are carried out to the correction type discrete cosine transform (MDCT) of different length.
On this basis, further parameter stereo coding is proposed again.Under this technology refers to and utilizes a road, mixing sound road is through monophony scrambler, and the basic framework of stereo parameter information additional a small amount of in addition, encodes to dual-channel audio data.Because stereo coding module and monophony coding module have employed the time-frequency conversion of different resolution, sound signal needs through time-frequency conversion process repeatedly, because this increasing the complicacy of encoding and decoding flow process entirety, too increase the time delay in encoding-decoding process simultaneously.
Therefore, for the problem that existing stereo coding/decoding method overall flow is too complicated, need a kind of new stereo coding/decoding method to reduce the complexity of overall flow.
Summary of the invention
For the problem that existing stereo coding/decoding method overall flow is too complicated, the invention provides a kind of stereo encoding method, described method comprises following steps:
Time-domain signal obtaining step, obtains the pairing sound channel time-domain signal of target audio;
Block length selects step, carries out block length select process thus obtain the time-frequency conversion block of pairing sound channel time-domain signal and corresponding block length control signal described pairing sound channel time-domain signal;
Time-frequency conversion step, carries out time-frequency conversion thus acquisition corresponding pairing vocal tract spectrum coefficient block based on described block length control signal to described time-frequency conversion block;
Stereo parameter coding step, carries out stereo parameter coding thus acquisition stereo parameter coded data block based on described block length control signal to described pairing vocal tract spectrum coefficient block;
Monophony treatment step, obtains monophony coded data block based on described block length control signal and described stereo parameter coded data block according to described pairing vocal tract spectrum coefficient block;
Encapsulation step, carries out code stream encapsulation by described monophony coded data block, described stereo parameter coded data block and described block length control signal thus obtains stereo coding packet.
In one embodiment, described stereo parameter coding step comprises following steps:
Stereo parameter extracts mode selecting step, selects corresponding stereo parameter to extract pattern and generate corresponding stereo parameter to extract mode flag according to the encoder bit rate of the stereo parameter part of described target audio and described block length control signal;
Stereo parameter extraction step, extracts pattern according to described stereo parameter from described pairing vocal tract spectrum coefficient block, extracts corresponding stereo parameter;
Stereo parameter quantization encoding step, carries out quantization encoding to described stereo parameter thus obtains the stereo parameter quantized, and then obtaining described stereo parameter coded data block.
In one embodiment, described stereo parameter is extracted pattern and is comprised common extraction pattern and superframe extraction pattern, wherein:
Under described common extraction pattern, a corresponding described stereo parameter coded data block of described monophony coded data block;
Under described superframe extraction pattern, the corresponding described stereo parameter coded data block of monophony coded data block described in multiple continuous print.
In one embodiment, described stereo parameter extraction mode selecting step comprises following steps:
Encoder bit rate analytical procedure, selects corresponding stereo parameter to extract pattern, when the encoder bit rate of described stereo parameter part is higher than selecting described common extraction pattern during particular value according to the encoder bit rate of described stereo parameter part;
Block length analytical procedure, when the encoder bit rate of described stereo parameter part is less than or equal to analyzing described block length control signal during particular value and selecting corresponding stereo parameter to extract pattern according to block length control signal analysis result.
In one embodiment, described stereo parameter is extracted mode selecting step and is also comprised lower mixed energy attenuation analytical procedure, analyzes the lower mixed energy attenuation of described pairing vocal tract spectrum coefficient block under different stereo parameter extraction pattern and selects corresponding stereo parameter to extract pattern based on analysis result.
In one embodiment, described monophony treatment step comprises following steps:
Mixed step under frequency spectrum, the stereo parameter based on described quantification carries out lower mixed process thus the monophony spectral coefficient block of acquisition correspondence to described pairing vocal tract spectrum coefficient block;
Monophony coding step, carries out monophony coded treatment based on described block length control signal to described monophony spectral coefficient block thus obtains described monophony coded data block.
In one embodiment, in described encapsulation step, described monophony coded data block and the described stereo parameter coded data block corresponding with described monophony coded data block, described block length control signal and described stereo parameter are extracted mode flag and carry out code stream encapsulation with certain format thus obtain described stereo coding packet.
Present invention also offers a kind of stereo decoding method, said method comprising the steps of:
Decapsulation step, carries out decapsulation process to described stereo coding packet thus obtains monophony coded data block, stereo parameter coded data block, block length control signal and stereo parameter extracting mode flag;
Monophony decoding step, carries out monophony decoding process thus acquisition monophony spectral coefficient block based on described block length control signal to described monophony coded data block;
Stereo parameter decoding step, extracts mode flag based on described block length control signal and stereo parameter, carries out decoding process thus obtain stereo parameter corresponding to described monophony spectral coefficient block to described stereo parameter coded data block;
Stereo treatment step, carries out stereo process thus acquisition pairing vocal tract spectrum coefficient block based on described block length control signal according to described monophony spectral coefficient block and described stereo parameter;
Time-frequency inverse transformation step, carries out time-frequency inverse transformation thus the pairing sound channel time-domain signal of acquisition target audio based on described block length control signal to described pairing vocal tract spectrum coefficient block.
In one embodiment, in described stereo parameter decoding step, extract mode flag according to described block length control signal and stereo parameter and select corresponding stereo parameter decoding schema.
In one embodiment, described stereo parameter decoding schema comprises common decoding schema and superframe decoding schema.
Compared with prior art, tool of the present invention has the following advantages:
Decoding method of the present invention provides the fusion framework compacter with core mono codec, multiplexing time-frequency conversion step, thus the calculated amount of having cut down because time-frequency conversion repeatedly brings and delay overhead;
Coding method of the present invention provides the stereo encoding method used in combination of different mode more flexibly, reduces the data volume of final coding result, thus ensure that difference stereo code check configuration under can obtain higher audio quality.
Further feature of the present invention or advantage will be set forth in the following description.Further, Partial Feature of the present invention or advantage will be become apparent by instructions, or be understood by implementing the present invention.Object of the present invention and certain advantages realize by step specifically noted in instructions, claims and accompanying drawing or obtain.
Accompanying drawing explanation
Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for instructions, with embodiments of the invention jointly for explaining the present invention, is not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is according to one embodiment of the invention coding process flow diagram;
Fig. 2 divides schematic diagram according to one embodiment of the invention time-frequency conversion block;
Fig. 3 extracts model selection process flow diagram according to one embodiment of the invention stereo parameter;
Fig. 4 is according to one embodiment of the invention encoded data structure sketch;
Fig. 5 is according to one embodiment of the invention stereo coding packet structure sketch;
Fig. 6 is according to one embodiment of the invention decoding process figure.
Embodiment
Embodiments of the present invention are described in detail below with reference to drawings and Examples, enforcement personnel of the present invention whereby can fully understand how application technology means solve technical matters in the present invention, and reach the implementation procedure of technique effect and specifically implement the present invention according to above-mentioned implementation procedure.It should be noted that, only otherwise form conflict, each embodiment in the present invention and each feature in each embodiment can be combined with each other, and the technical scheme formed is all within protection scope of the present invention.
The present invention proposes a kind of stereo coding/decoding method.Next the encoding and decoding idiographic flow of the inventive method is specifically described based on process flow diagram.Step shown in the process flow diagram of accompanying drawing can perform in the computer system comprising such as one group of computer executable instructions.Although show the logical order of each step in flow charts, in some cases, can be different from the step shown or described by order execution herein.
First be described for cataloged procedure.As shown in Figure 1, first perform step S100, time-domain signal obtaining step, obtain the pairing sound channel time-domain signal of target audio.First coding method of the present invention will carry out time-frequency conversion to the time-domain signal of audio frequency, and time-domain signal is converted into corresponding spectral coefficient.Pairing sound channel time-domain signal is the right set of numerous signal sampling points.Due to when carrying out coded treatment to audio frequency, usually in units of frame, sound signal is processed.Therefore, before time-frequency conversion, first all time-domain signals are divided into multiple continuous print time-domain signal frame.Definition M is frame length, and a continuous N signal sampling point is to being a frame.
In the present embodiment, pairing sound channel is left and right acoustic channels, the time-domain signal block of each code period process superframe length.Superframe length is the integral multiple of frame length, and definition N is superframe length parameter, and the time-domain signal block of a superframe length comprises N number of time-domain signal frame, i.e. M × N number of signal sampling point pair.Namely step S100 reads in each M of left and right acoustic channels × N number of sampling point (M × N number of sampling point to) at every turn.
In order to obtain higher compression coding gain, coding method of the present invention provides sound signal optimal time frequency resolution by adopting the time-frequency conversion of adaptive block length, and this just needs, according to the actual conditions of sound signal, a time-domain signal frame is divided into one or more time-frequency conversion block.Therefore in the present embodiment, perform step S110 after step sloo, block length selects step, carries out block length select process thus obtain the time-frequency conversion block of pairing sound channel time-domain signal and corresponding block length control signal pairing sound channel time-domain signal.
In the present embodiment, short block combination in the long block combination of interframe and frame is supported to the process of sound signal.
Frame frequency spectral coefficient data (namely with the data block that M sampling point is formed spectral coefficient) after short block combination refers to time-frequency conversion in frame can be that multiple m (m is less than M) individual sampling point is to the combined result of the data block that spectral coefficient is formed.Can be such as two M/2 sampling points to the combination of the data block that spectral coefficient is formed, or 4 M/4 sampling points are to the combination of the data block that spectral coefficient is formed.In this case, the sub-band division of the spectral coefficient after time-frequency conversion and stereo parameter are extracted all correspondences and are carried out in each time-frequency conversion block.For different time-frequency conversion block lengths, the mode that selection divides also can be different, comprise the number of spectral sub-bands, or the number of each sub-band coefficients.
The combination of interframe long block refers to and allows the spectral coefficient of continuous 2 (or multiple) M point signal frames to combine, and forms a matrix of coefficients.But require that these spectral coefficients are all adopt the time-frequency conversion of M point to obtain.In this case, the sub-band division of spectral coefficient and stereo parameter are extracted and all correspondingly to be carried out in matrix of coefficients.
Based on short block combination in frame and the long block combination of interframe, in step s 110, each time-domain signal frame (M sampling point to) is divided into a time-frequency conversion block (M sampling point to) or multiple time-frequency conversion block (m sampling point to).Suppose that the time-domain signal block of a superframe length comprises 3M signal sampling point to (superframe length N=3).Then as shown in Figure 2, the time-domain signal block 210 of superframe length comprises signal frame 211,212 and 213, and each signal frame (211,212 or 213) comprises M signal sampling point pair.Actual conditions according to sound signal do Further Division to each signal frame (211,212 or 213).First carry out 1/2 time to the time-domain signal of each signal frame (M signal sampling point to) to mix, suppose that the signal of left and right acoustic channels is respectively X
land X
r, lower mixed result X
mcomputing formula as follows:
The M obtained a lower mixed sampling point is analyzed, thus obtain a block length control signal corresponding with time-domain signal frame, the block length information (can be an independent M point block length, or a series of M/2, M/4, M/8 etc. more short block is long) of each time-frequency conversion block when this block length control signal is used for representing that this signal frame carries out time-frequency conversion.And then according to the block length control signal of correspondence, one or more time-frequency conversion block is divided into left and right acoustic channels time-domain signal frame.
Division result can be the one in Fig. 2 in time-frequency conversion block division result 220,230 and 240.Situation beyond frequency transform block division result shown when certain division result also can be Fig. 2.
As shown in Fig. 2 time-frequency conversion block division result 220, signal frame 211 is divided into and comprises the right time-frequency conversion block 221 of M/2 signal sampling point, comprise the right time-frequency conversion block 222 and 223 of M/4 signal sampling point; Signal frame 212 is divided into and comprises the right time-frequency conversion block 224 of M signal sampling point; Signal frame 213 is divided into and comprises the right time-frequency conversion block 225 of M signal sampling point.
As shown in Fig. 2 time-frequency conversion block division result 230, signal frame 211 is divided into and comprises the right time-frequency conversion block 231 and 232 of M/2 signal sampling point; Signal frame 212 is divided into and comprises the right time-frequency conversion block 233 and 234 of M/2 signal sampling point; Signal frame 213 is divided into and comprises the right time-frequency conversion block 235 and 236 of M/2 signal sampling point.
As shown in Fig. 2 time-frequency conversion block division result 240, signal frame 211 is divided into and comprises the right time-frequency conversion block 241 of M signal sampling point; Signal frame 212 is divided into and comprises the right time-frequency conversion block 242 of M signal sampling point; Signal frame 213 is divided into and comprises the right time-frequency conversion block 243 of M signal sampling point.
As shown in Fig. 2 time-frequency conversion block division result 250, signal frame 211 is divided into and comprises the right time-frequency conversion block 251 of M signal sampling point; Signal frame 212 is divided into and comprises the right time-frequency conversion block 252 of M/2 signal sampling point, comprise the right time-frequency conversion block 253 and 254 of M/4 signal sampling point; Signal frame 213 is divided into and comprises the right time-frequency conversion block 255 of M signal sampling point.
Next just can perform step S120 as shown in Figure 1, time-frequency conversion step, based on block length control signal, time-frequency conversion be carried out thus acquisition corresponding pairing vocal tract spectrum coefficient block to time-frequency conversion block; Each time-frequency conversion block generates a corresponding pairing vocal tract spectrum coefficient block after time-frequency conversion.
Just can encode to pairing vocal tract spectrum coefficient block after step S120.In the present embodiment, coding step comprises stereo parameter coding step, based on block length control signal, pairing vocal tract spectrum coefficient block is carried out to stereo parameter coding thus is obtained stereo parameter coded data block.Stereo parameter coding step comprises step S140, stereo parameter extraction step, extracts stereo parameter based on block length control signal from pairing vocal tract spectrum coefficient block.
For effectively reducing encoder bit rate, coding method of the present invention proposes superframe and extracts pattern and common extraction pattern two kinds of stereo parameter extraction patterns.Therefore before execution step S140, need to perform step S142, stereo parameter extracts mode selecting step, selects corresponding stereo parameter to extract pattern and generate corresponding stereo parameter to extract mode flag according to the encoder bit rate of the stereo parameter part of target audio and block length control signal.In the present embodiment, stereo parameter is extracted mode flag and whether be have employed superframe extraction pattern for identifying in the present encoding cycle.Definition stereo parameter extracts the superframe extraction mode identifier sflag that mode flag is 1bit.Sflag=1 represents that employing superframe extracts mode treatment; Sflag=0 represents the common extraction mode treatment of employing.
Under common extraction pattern, first respectively frequency domain sub-band division is carried out to the spectral coefficient block of pairing sound channel Zhong Mei road sound channel according to block length control signal, obtain several spectral coefficient vectors; Then spectral coefficient vector corresponding to pairing sound channel is utilized to carry out stereo parameter extraction.Under common coding mode, each pairing vocal tract spectrum coefficient block generates one group of stereo parameter.
Under superframe extraction pattern, the sampling point number that the support of stereo coding module processes at every turn is the situation of the integral multiple n (2≤n≤N) of the frame length of core encoder resume module.First n M point spectral coefficient block of pairing sound channel Zhong Mei road sound channel is formed a two-dimentional spectral coefficient matrix; Then the spectral coefficient matrix of every road sound channel is carried out to the sub-band division in frequency dimension, obtain several spectral coefficient submatrixs; Then utilize pairing sound channel often to organize corresponding spectral coefficient submatrix and carry out stereo parameter extraction.Namely, under superframe extraction pattern, the pairing vocal tract spectrum coefficient block that n to comprise M sampling point right generates one group of stereo parameter.Compared to common extraction pattern, the data volume of the stereo parameter of generation reduces greatly.The detailed implementation of step S142 as shown in Figure 3, first performs step S310, encoder bit rate analytical procedure, selects corresponding stereo parameter to extract pattern according to the encoder bit rate of stereo parameter part.In the present embodiment, when the encoder bit rate of stereo parameter part is higher than certain specific threshold, common extraction pattern is adopted to carry out stereo parameter extraction to pairing vocal tract spectrum coefficient block.Usually, the encoder bit rate of stereo parameter part is the preset parameter set before coding, sets the specific threshold of the encoder bit rate of stereo parameter part according to the actual coding situation of audio frequency in advance.In the present embodiment, setting this specific threshold is 12kbps.
When the encoder bit rate of stereo parameter part is less than or equal to above-mentioned specific threshold, just can perform step S320, block length analytical procedure, analyze block length control signal and select corresponding stereo parameter to extract pattern according to block length control signal analysis result.
Based on the description of the above-mentioned processing mode to super frame mode, scrambler reads in the time-domain signal (M × N number of signal sampling point to) of a superframe length at every turn, after carrying out time-frequency conversion, when there is continuous n (n >=2) and organizing the pairing vocal tract spectrum coefficient block of M point, superframe just can be adopted to extract pattern.Namely when the time-domain signal frame in the time-domain signal to a superframe length divides, if when division result comprises the time-frequency conversion block of continuous print n M point length, the stereo parameter that the corresponding pairing vocal tract spectrum coefficient block that just can generate it is done under superframe extraction pattern is extracted, thus obtains one group of common stereo parameter.Be understood that, when the time-domain signal block of a superframe length only comprises one signal frame (N=1), superframe cannot be adopted to extract pattern stereo parameter extraction is carried out to it.
In the present embodiment, it is block length analysis condition that definition exists continuous n M point block length control signal, when the condition is not satisfied, namely discontinuous when the time-frequency conversion block of M point, or the length of time-frequency conversion block is when being less than M, then perform step S301, select common extraction pattern for above-mentioned time-frequency conversion block.
When extracting pattern owing to adopting common extraction pattern and superframe, be different to the lower mixed result (the monophony spectral coefficient namely generated) of pairing sound channel.In order to obtain more desirable coding result, in the present embodiment, stereo parameter is extracted mode selecting step and is also comprised lower mixed energy attenuation analytical procedure (step S330), analyzes the lower mixed energy attenuation of pairing vocal tract spectrum coefficient block under different stereo parameter extraction pattern and selects corresponding stereo parameter to extract pattern based on analysis result.
In the present embodiment, judge to meet block length analysis condition in step S320, when namely there is the pairing vocal tract spectrum coefficient block of continuous n group (2≤n≤N) M point, perform step S330.First respectively carrying out lower mixed process based on superframe extraction pattern and the pairing vocal tract spectrum coefficient block of common extraction pattern to continuous n group (2≤n≤N) M point, taking superframe to extract pattern according to determining whether the energy attenuation analysis result of the lower mixed process under two kinds of patterns.Specific implementation step is as follows:
Suppose that the pairing vocal tract spectrum coefficient block of continuous n group (2≤n≤N) M point is expressed as L (i, j) and R (i, j), wherein i=0 ..., M-1, j=0 ..., n-1.Adopt superframe extract pattern and common extraction pattern obtain corresponding lower mixing pedigree several piece be expressed as S (i, j) and
calculate its lower mixed energy
E
S＝∑
i,jS
2(i,j) (2)
And off-energy
Wherein: E
sfor superframe extracts lower mixed energy corresponding to pattern;
for the lower mixed energy that common extraction pattern is corresponding; E
dfor superframe extracts off-energy corresponding to pattern;
for the off-energy that common extraction pattern is corresponding.
When meeting the following conditions for the moment, so this n group spectral coefficient block adopts superframe to extract pattern; Otherwise, adopt common extraction pattern.
A) E
d/ E
s<T
1(threshold value T
1can be any value between 0 ~ 0.5 in theory, get 0.05 in the present embodiment);
B)
(threshold value T
2can be in theory be greater than 1.0 any value.Get 5.0 in the present embodiment).
Here it is pointed out that the object of step S330 is to obtain the more desirable coding result of coding quality, consider calculated amount or other practical operation factors, step S330 can omit.Namely in another embodiment of the invention, stereo parameter extraction mode selecting step only comprises encoder bit rate analytical procedure (S310) and block length analytical procedure (S320).
Next for the time-frequency conversion block division result 220,230 and 240 shown in Fig. 2, concrete description stereo parameter extracts the selection result of mode selecting step.Suppose that all lower mixed energy attenuatioies are analyzed all satisfy condition (omitting lower mixed energy attenuation analytical procedure in other words).
For time-frequency conversion block division result 220, common extraction pattern is adopted to carry out stereo parameter extraction to the pairing vocal tract spectrum coefficient block that time-frequency conversion block 221,222 and 223 generates; Superframe extraction pattern is adopted to carry out stereo parameter extraction to the pairing vocal tract spectrum coefficient block that time-frequency conversion block 224 and 225 generates.
For time-frequency conversion block division result 230, common extraction pattern is adopted to carry out stereo parameter extraction to the pairing vocal tract spectrum coefficient block that time-frequency conversion block 231,232,233,234,235 and 236 generates.
For time-frequency conversion block division result 240, superframe extraction pattern is adopted to carry out stereo parameter extraction to the pairing vocal tract spectrum coefficient block that time-frequency conversion block 241,242 and 243 generates.
For time-frequency conversion block division result 250, common extraction pattern is adopted to carry out stereo parameter extraction to the pairing vocal tract spectrum coefficient block that time-frequency conversion block 251,252,253,254 and 255 generates.
After stereo parameter is extracted, step S141 as shown in Figure 1 can be performed, quantization encoding step, quantization encoding is carried out to stereo parameter, thus one group of stereo parameter quantized, and then obtain stereo parameter coded data block
In the present embodiment, coding step also comprises monophony treatment step, obtains monophony coded data block based on block length control signal according to pairing vocal tract spectrum coefficient block and the stereo parameter after quantizing.In this step, first perform step S130, mixed step under frequency spectrum, the stereo parameter based on the quantification obtained in step S141 is carried out lower mixed process to pairing vocal tract spectrum coefficient block thus obtains corresponding monophony spectral coefficient block.
Then perform step S131, monophony coding step, monophony coded treatment is carried out to monophony spectral coefficient block thus obtains monophony coded data block.Using all monophony spectral coefficient blocks corresponding to N number of time-domain signal frame and corresponding block length control signal as input, send into core encoder successively, obtain corresponding monophony coded data block.Each pairing vocal tract spectrum coefficient block can generate a monophony coded data block.Be understood that, if what adopt in step S140 is common extraction pattern, then the corresponding stereo parameter coded data block of each monophony coded data block.If what adopt in step S140 is that superframe extracts pattern, then the corresponding stereo parameter coded data block of multiple monophony coded data block.Finally just step S150 can be performed, encapsulation step, extracts mode flag and carries out code stream encapsulation with certain format thus obtain stereo coding packet by monophony coded data block and the stereo parameter coded data block corresponding with monophony coded data block, block length control signal and stereo parameter.
In the present embodiment, each code period is read in the time-domain signal block of a superframe length and carry out encapsulation step at the end of each code period.Therefore the information of the time-domain signal block of a superframe length is comprised in each stereo coding packet.Certainly, also after processing the time-domain signal block of multiple superframe length, disposable encapsulation can be done to the result of the time-domain signal block of multiple superframe length.
The coded data packet each code period obtained combines the final coding result that namely can obtain target audio file successively according to specific format.As shown in Figure 4, what form 400 was described is the data structure of a certain sound signal after coding method of the present invention is encoded.It is by a head information packet and multiple stereo coding packet (coded data packet 1,2,3 ...) form.
For stereo coding packet 1, suppose that the raw information of stereo coding packet 1 is the time domain time-frequency conversion block 210 of superframe length as shown in Figure 2.For time-frequency conversion block division result 220, its data structure schematic diagram finally generating stereo coding packet is as shown in Fig. 5 neutral body sound encoder packet 501.Stereo coding packet 501 comprises sound channel and extracts mode flag 1, multiple block length control signal, multiple monophony coded data block and multiple stereo coding data block to identifier 1, stereo parameter.
In stereo coding packet 501, signal frame 211 in block length control signal 1-1 corresponding diagram 2, i.e. time-frequency transform block 221,222 and 223; Signal frame 212 in block length control signal 2-1 corresponding diagram 2, i.e. time-frequency transform block 224; Signal frame 213 in block length control signal 3-1 corresponding diagram 2, i.e. time-frequency transform block 225.Time-frequency conversion block 221 in monophony coded data block 1-1 and stereo parameter coded data block 1-1 corresponding diagram 2; Time-frequency conversion block 222 in monophony coded data block 2-1 and stereo parameter coded data block 2-1 corresponding diagram 2; Time-frequency conversion block 223 in monophony coded data block 3-1 and stereo parameter coded data block 3-1 corresponding diagram 2; Time-frequency conversion block 224 in monophony coded data block 4-1 corresponding diagram 2; Time-frequency conversion block 225 in monophony coded data block 5-1 corresponding diagram 2.Time-frequency conversion block 224 and 225 in stereo parameter coded data block 4-1 corresponding diagram 2.It is sflag=1 that stereo parameter extracts mode flag 1.
For frequency transform block division result 240, the data block that its coded data packet generated comprises is as shown in Fig. 4 data structure schematic diagram 502.Stereo coding packet 502 comprises sound channel and extracts mode flag 2, multiple block length control signal, multiple monophony coded data block and a stereo coding data block to identifier 2, stereo parameter.Wherein, signal frame 211, i.e. time-frequency transform block 241 in block length control signal 1-2 and monophony coded data block 1-2 corresponding diagram 2; Signal frame 212, i.e. time-frequency transform block 242 in block length control signal 2-2 and monophony coded data block 2-2 corresponding diagram 2; Signal frame 213, i.e. time-frequency transform block 243 in block length control signal 3-2 and monophony coded data block 3-2 corresponding diagram 2.Time-frequency conversion block 241,242 and 243 in stereo parameter coded data block 1-2 corresponding diagram 2.It is sflag=1 that stereo parameter extracts mode flag 2.
For frequency transform block division result 250, the data block that its coded data packet generated comprises is as shown in Fig. 4 data structure schematic diagram 503.Stereo coding packet 503 comprises sound channel and extracts mode flag 3, multiple block length control signal, multiple monophony coded data block and multiple stereo coding data block to identifier 3, stereo parameter.Wherein, signal frame 211 in block length control signal 1-3 corresponding diagram 2, i.e. time-frequency transform block 251; Signal frame 212 in block length control signal 2-3 corresponding diagram 2, i.e. time-frequency transform block 252,253 and 254; Signal frame 213 in block length control signal 3-3 corresponding diagram 2, i.e. time-frequency transform block 255.Time-frequency conversion block 251 in monophony coded data block 1-3 and stereo parameter coded data block 1-3 corresponding diagram 2; Time-frequency conversion block 252 in monophony coded data block 2-3 and stereo parameter coded data block 2-3 corresponding diagram 2; Time-frequency conversion block 253 in monophony coded data block 3-3 and stereo parameter coded data block 3-3 corresponding diagram 2; Time-frequency conversion block 254 in monophony coded data block 4-3 and stereo parameter coded data block 4-3 corresponding diagram 2; Time-frequency conversion block 255 in monophony coded data block 5-3 and stereo parameter coded data block 5-3 corresponding diagram 2.It is sflag=0 that stereo parameter extracts mode flag 2.
Correlation data structural representation 501,502 and 503 can be found out, adopts superframe to extract schema extraction stereo parameter, substantially reduces the data volume of stereo coding packet.Coding method of the present invention provides superframe more flexibly and extracts the stereo encoding method that pattern and common extraction mode mixture use, and ensure that can obtain equal audio quality with lower stereo code check.
What it is pointed out that data structure schematic diagram 501,502 and 503 in Fig. 5 describes is the part of the data block that stereo coding packet comprises.In actual coding, the arrangement mode of each data block is determined by the encapsulation format of coded data packet, and according to the encapsulation format of coded data packet and other coding requirements of reality, the data block of other types and content in stereo coding packet, can be comprised.
Next be described for coding/decoding method of the present invention.Stereo decoding method of the present invention is the method that the coded data packet generated for stereo encoding method of the present invention carries out decoding.As shown in Figure 6, first perform step S600, decapsulation step, decapsulation process is carried out to coded data packet thus obtains monophony coded data block, stereo parameter coded data block and block length control signal.
Next just can perform step S610 monophony decoding step, based on block length control signal, monophony decoding process be carried out thus acquisition monophony spectral coefficient block to monophony coded data block.
While execution step S610, perform step S620, stereo parameter decoding step, carry out decoding process thus obtaining stereo parameter corresponding to monophony spectral coefficient block to stereo parameter coding data block based on block length control signal.
Common extraction pattern and the superframe of corresponding coding method neutral body sound parameter extraction pattern of the present invention extract pattern, and coding/decoding method of the present invention comprises common decoding schema and superframe decoding schema two kinds of stereo parameter decoding schemas.
In step S620, extract mode flag according to block length control signal and stereo parameter and judge that current decoded data frame obtains the mode of corresponding stereo parameter, thus choose corresponding stereo parameter decoding schema.
The step S600 selected based on common decoding schema and superframe decoding schema, S610 and S620 specific operation process are as follows:
In step S600, first determine superframe length Parameter N and frame length M; Then carry out decapsulation process to each decoded data bag thus obtain stereo parameter extracting mode flag (superframe extracts mode identifier sflag); Then, decoded data bag is split into N number of decoded data frame (comprising the information of a time-domain signal frame in each decoded data frame) in units of frame.
In step S610, monophony coded data block corresponding for current decoded data frame and block length control signal are sent into core decoder, carry out inverse quantization and obtain corresponding monophony spectral coefficient block.
In step S620, if the block length control signal of current decoded data frame and superframe extract mode identifier sflag meet following 3 conditions, carry out step S620.1; Otherwise carry out step S620.2.
A. superframe extracts mode identifier sflag=1;
B. current decoded data frame is not the first frame in current decoding treatment cycle;
C. the block length control signal of current decoded data frame and a upper decoded data frame all represent the time-frequency conversion adopting M point.
Step S620.1: current decoded data frame adopts the one group stereo parameter identical with a upper decoded data frame.
Step S620.2: extract stereo parameter coded data block from current decoded data frame, inverse quantization obtains stereo parameter.
Be not difficult to find, as N=1, namely each decoded data bag only comprises the information of a time-domain signal frame, and therefore decoded data bag can only adopt step S620.2 to obtain corresponding stereo parameter.When N >=2, each decoded data frame in decoded data bag needs to carry out judgement selection and carries out step S620.1 or step S620.2.
The basis of step S610 and S620 just can perform step S630, stereo treatment step, carry out stereo process thus acquisition pairing vocal tract spectrum coefficient block based on block length control signal according to monophony spectral coefficient block and stereo parameter.In step S630, stereo mixed process is carried out to the M point spectral coefficient of current decoded data frame and the stereo parameter of correspondence, obtains pairing sound channel M point spectral coefficient separately.
Next just can perform step S640, time-frequency inverse transformation step, based on block length control signal time-frequency inverse transformation be carried out to pairing vocal tract spectrum coefficient thus the time domain reconstruction signal of the pairing sound channel obtaining target audio M point separately.Finally perform step S650, the time domain reconstruction signal obtained in step S640 is exported.
Stereo coding/decoding method of the present invention, selects identical block length selection mode multiplexing with core encoder and time-frequency conversion step in whole coding/decoding process.During coding, using the monophony frequency spectrum data of acquisition after stereo parameter coding as input, directly send into core encoder and carry out quantization encoding; During decoding, export monophony frequency spectrum data as input from core decoder, send into stereo decoding processing module.Which provides the fusion framework compacter with core mono scrambler, thus the calculated amount of having cut down because time-frequency conversion repeatedly brings and delay overhead.
Coding method of the present invention also proposes this stereo parameter of superframe extraction pattern extraction pattern based on the mode of the long block combination of interframe, effectively reduces final stereo coding data volume.Coding method of the present invention is by extracting pattern and common extraction pattern used in combination based on short block combination in frame to superframe, provide stereo encoding method more flexibly, ensure that difference stereo code check configuration under can obtain higher audio quality.
Next the execution flow process of stereo coding/decoding method of the present invention is described by a concrete application example.For superframe length Parameter N=2, namely scrambler reads in the signal sampling point pair of 2M at every turn.Stereo coding/decoding module adopts the greatly relevant decoding method rotated, and block length selects the optional Output rusults of module to comprise the long block structure of M point and the short block unitized construction of 8 M/8 points, and time-frequency conversion module adopts MDCT conversion.
During coding, first 1/2 time is carried out to the 2M point time-domain signal of left and right acoustic channels and mix, and be divided into the time-domain signal frame of 2 M points, send into block length successively and select module.Complete corresponding MDCT according to block length Output rusults to left and right acoustic channels to convert, obtain spectral coefficient.
A) when 2 frame block length Output rusults all adopt long piece of M point, the spectral coefficient of 2 M points of every road sound channel is combined into the matrix of coefficients of M × 2, represents with L and R respectively.Matrix of coefficients L and R is carried out to the partition of frequency dimension, divide number and be expressed as N
l.Then to often organizing coefficient sub-block L
kand R
k, (k=0 ..., N
l-1) a greatly relevant rotation angle θ is extracted
k:
Wherein θ
0for relevant rotation angle parameter, and
Here, the definition of inner product expands to two-dimensional matrix space:
Wherein: l (i, j) and r (i, j) represents coefficient sub-block L respectively
kand R
kall elements.
Quantization encoding is carried out to all rotation angle, obtains a stereo parameter coded data block.Meanwhile, the rotation angle after quantizing is utilized
to often organizing coefficient sub-block L
kand R
kgreatly to be correlated with rotational transform, to obtain two coefficient sub-blocks after rotational transform
with
To coefficient sub-block
with
carry out lower mixed again, obtain corresponding lower mixed coefficient sub-block
Reconfigure all lower mixed coefficient sub-blocks, obtain the lower mixed matrix of coefficients of M × 2.Successively using each column vector of this matrix as input, send into core encoder encode, obtain 2 monophony coded data blocks.
During code stream encapsulation, first monophony coded data block and corresponding block length control signal, can form a coded frame data jointly with stereo parameter coded data block; Second monophony coded data block and corresponding block length control signal form a coded frame data.Finally two coded frame datas and superframe are extracted mode identifier sflag (here sflag=1) and form a coded data packet.
B) when when 2 frame block length Output rusults, at least one adopts 8 M/8 point short blocks to combine, so the greatly relevant rotation angle of each time-frequency conversion block (M point or M/8 point) complete independently is extracted, and the greatly relevant rotational transform of sub-band coefficients vector and under sneak out journey.
For the spectral coefficient block adopting the lower mixing spectral coefficient of continuous 8 M/8 point short blocks can be reassembled into 1 M point, send into core encoder together with the block length control signal of correspondence, obtain 1 monophony coded data block; 8 groups of greatly relevant rotation angle that correspondence obtains, after quantization encoding, also can be combined into 1 stereo parameter coded data block.
For adopting the lower mixing spectral coefficient of long piece of M point directly can send into core encoder together with corresponding block length control signal, obtain 1 monophony coded data block.1 group of greatly relevant rotation angle that correspondence obtains, after quantization encoding, also can be combined into 1 stereo parameter coded data block.
During code stream encapsulation, each monophony coded data block, can form a coded frame data jointly with corresponding block length control signal and stereo parameter coded data block.Finally two coded frame datas and superframe are extracted mode identifier sflag (here sflag=0) and form a coded data packet.
During decoding, first superframe is extracted to each decoded data bag and extracts mode identifier sflag; Then, decoded data bag is splitted into 2 decoded data frames.
Block length control signal, monophony coded data block and stereo parameter coded data block are divided into further to the 1st decoded data frame.According to block length control signal, monophony coded data block and stereo parameter coded data block are carried out inverse quantization decoding respectively, obtain M point spectral coefficient and corresponding greatly relevant rotation angle.Then, in each time-frequency conversion block, the upper mixed conversion of complete independently sub-band coefficients vector and IMDCT process, reconstruct the 1st group of M point time-domain signal of left and right acoustic channels.
To the 2nd decoded data frame, point following 2 kinds of situations are carried out:
(1) if sflag=1 time, so only have monophony coded data block to carry out inverse quantization, obtain M point spectral coefficient, greatly relevant rotation angle is then identical with previous frame.
(2) if sflag=0 time, this decoded data frame adopt and the 1st stereo decoding method that decoded data frame is identical, reconstruct the 2nd group of M point time-domain signal of left and right acoustic channels.
Describe according to example, the cataloged procedure of this stereo coding/decoding method only carries out a MDCT conversion to every road sound channel, and decode procedure only carries out once oppositely Modified Discrete Cosine Tr ansform (IMDCT) conversion.Compare classic method, this stereo coding/decoding method, owing to providing the time frequency resolution consistent with core coding/decoding module to input signal, because omitted herein the pilot process using time-domain signal as transition, reduces encoding and decoding complexity; For using MDCT as the coding/decoding system of time-frequency conversion, also reduce encoding and decoding time delay simultaneously.In addition, due to the multiplexing stereo parameter coded data block of the monophony coded data block of partial continuous, effectively encoder bit rate can be reduced, optimally the encoder bit rate of stereo parameter part compressible about 50%.
Although embodiment disclosed in this invention is as above, the embodiment that described content just adopts for the ease of understanding the present invention, and be not used to limit the present invention.Method of the present invention also can have other various embodiments.When not deviating from essence of the present invention, those of ordinary skill in the art are when making various corresponding change or distortion according to the present invention, but these change accordingly or are out of shape the protection domain that all should belong to claim of the present invention.
Claims (10)
1. a stereo encoding method, is characterized in that, described method comprises following steps:
Time-domain signal obtaining step, obtains the pairing sound channel time-domain signal of target audio;
Block length selects step, carries out block length select process thus obtain the time-frequency conversion block of pairing sound channel time-domain signal and corresponding block length control signal described pairing sound channel time-domain signal;
Time-frequency conversion step, carries out time-frequency conversion thus acquisition corresponding pairing vocal tract spectrum coefficient block based on described block length control signal to described time-frequency conversion block;
Stereo parameter coding step, carries out stereo parameter coding thus acquisition stereo parameter coded data block based on described block length control signal to described pairing vocal tract spectrum coefficient block;
Monophony treatment step, obtains monophony coded data block based on described block length control signal and described stereo parameter coded data block according to described pairing vocal tract spectrum coefficient block;
Encapsulation step, carries out code stream encapsulation by described monophony coded data block, described stereo parameter coded data block and described block length control signal thus obtains stereo coding packet.
2. the method for claim 1, is characterized in that, described stereo parameter coding step comprises following steps:
Stereo parameter extracts mode selecting step, selects corresponding stereo parameter to extract pattern and generate corresponding stereo parameter to extract mode flag according to the encoder bit rate of the stereo parameter part of described target audio and described block length control signal;
Stereo parameter extraction step, extracts pattern according to described stereo parameter from described pairing vocal tract spectrum coefficient block, extracts corresponding stereo parameter;
Stereo parameter quantization encoding step, carries out quantization encoding to described stereo parameter thus obtains the stereo parameter quantized, and then obtaining described stereo parameter coded data block.
3. method as claimed in claim 2, is characterized in that, described stereo parameter is extracted pattern and comprised common extraction pattern and superframe extraction pattern, wherein:
Under described common extraction pattern, a corresponding described stereo parameter coded data block of described monophony coded data block;
Under described superframe extraction pattern, the corresponding described stereo parameter coded data block of monophony coded data block described in multiple continuous print.
4. method as claimed in claim 2, is characterized in that, described stereo parameter is extracted mode selecting step and comprised following steps:
Encoder bit rate analytical procedure, selects corresponding stereo parameter to extract pattern, when the encoder bit rate of described stereo parameter part is higher than selecting described common extraction pattern during particular value according to the encoder bit rate of described stereo parameter part;
Block length analytical procedure, when the encoder bit rate of described stereo parameter part is less than or equal to analyzing described block length control signal during particular value and selecting corresponding stereo parameter to extract pattern according to block length control signal analysis result.
5. method as claimed in claim 4, it is characterized in that, described stereo parameter is extracted mode selecting step and is also comprised lower mixed energy attenuation analytical procedure, analyzes the lower mixed energy attenuation of described pairing vocal tract spectrum coefficient block under different stereo parameter extraction pattern and selects corresponding stereo parameter to extract pattern based on analysis result.
6. method as claimed in claim 2, it is characterized in that, described monophony treatment step comprises following steps:
Mixed step under frequency spectrum, the stereo parameter based on described quantification carries out lower mixed process thus the monophony spectral coefficient block of acquisition correspondence to described pairing vocal tract spectrum coefficient block;
Monophony coding step, carries out monophony coded treatment based on described block length control signal to described monophony spectral coefficient block thus obtains described monophony coded data block.
7. method as claimed in claim 2, it is characterized in that, in described encapsulation step, described monophony coded data block and the described stereo parameter coded data block corresponding with described monophony coded data block, described block length control signal and described stereo parameter are extracted mode flag and carry out code stream encapsulation with certain format thus obtain described stereo coding packet.
8., for a stereo decoding method for the stereo coding packet of the method generation such as according to any one of claim 1-7, it is characterized in that, said method comprising the steps of:
Decapsulation step, carries out decapsulation process to described stereo coding packet thus obtains monophony coded data block, stereo parameter coded data block, block length control signal and stereo parameter extracting mode flag;
Monophony decoding step, carries out monophony decoding process thus acquisition monophony spectral coefficient block based on described block length control signal to described monophony coded data block;
Stereo parameter decoding step, extracts mode flag based on described block length control signal and stereo parameter, carries out decoding process thus obtain stereo parameter corresponding to described monophony spectral coefficient block to described stereo parameter coded data block;
Stereo treatment step, carries out stereo process thus acquisition pairing vocal tract spectrum coefficient block based on described block length control signal according to described monophony spectral coefficient block and described stereo parameter;
Time-frequency inverse transformation step, carries out time-frequency inverse transformation thus the pairing sound channel time-domain signal of acquisition target audio based on described block length control signal to described pairing vocal tract spectrum coefficient block.
9. method as claimed in claim 8, is characterized in that, in described stereo parameter decoding step, extracts mode flag select corresponding stereo parameter decoding schema according to described block length control signal and stereo parameter.
10. method as claimed in claim 9, it is characterized in that, described stereo parameter decoding schema comprises common decoding schema and superframe decoding schema.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410573759.9A CN104347077B (en) | 2014-10-23 | 2014-10-23 | A kind of stereo coding/decoding method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410573759.9A CN104347077B (en) | 2014-10-23 | 2014-10-23 | A kind of stereo coding/decoding method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104347077A true CN104347077A (en) | 2015-02-11 |
| CN104347077B CN104347077B (en) | 2018-01-16 |
Family
ID=52502545
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410573759.9A Active CN104347077B (en) | 2014-10-23 | 2014-10-23 | A kind of stereo coding/decoding method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104347077B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019029736A1 (en) * | 2017-08-10 | 2019-02-14 | 华为技术有限公司 | Time-domain stereo coding and decoding method and related product |
| WO2021000724A1 (en) * | 2019-06-29 | 2021-01-07 | 华为技术有限公司 | Stereo coding method and device, and stereo decoding method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1914668A (en) * | 2004-01-28 | 2007-02-14 | 皇家飞利浦电子股份有限公司 | Method and apparatus for time scaling of a signal |
| CN101128866A (en) * | 2005-02-23 | 2008-02-20 | 艾利森电话股份有限公司 | Optimized fidelity and reduced signaling in multi-channel audio encoding |
| CN101202043A (en) * | 2007-12-28 | 2008-06-18 | 清华大学 | Method and system for encoding and decoding audio signal |
| CN101604983A (en) * | 2008-06-12 | 2009-12-16 | 华为技术有限公司 | Coding and decoding device, system and method thereof |
| CN102428513A (en) * | 2009-03-18 | 2012-04-25 | 三星电子株式会社 | Apparatus And Method For Encoding/Decoding A Multichannel Signal |
-
2014
- 2014-10-23 CN CN201410573759.9A patent/CN104347077B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1914668A (en) * | 2004-01-28 | 2007-02-14 | 皇家飞利浦电子股份有限公司 | Method and apparatus for time scaling of a signal |
| US20090192804A1 (en) * | 2004-01-28 | 2009-07-30 | Koninklijke Philips Electronic, N.V. | Method and apparatus for time scaling of a signal |
| CN101128866A (en) * | 2005-02-23 | 2008-02-20 | 艾利森电话股份有限公司 | Optimized fidelity and reduced signaling in multi-channel audio encoding |
| CN101202043A (en) * | 2007-12-28 | 2008-06-18 | 清华大学 | Method and system for encoding and decoding audio signal |
| CN101604983A (en) * | 2008-06-12 | 2009-12-16 | 华为技术有限公司 | Coding and decoding device, system and method thereof |
| CN102428513A (en) * | 2009-03-18 | 2012-04-25 | 三星电子株式会社 | Apparatus And Method For Encoding/Decoding A Multichannel Signal |
Non-Patent Citations (1)
| Title |
|---|
| 张树华: ""MDCT域信号处理及其在立体声编码中的应用"", 《中国博士学位论文全文数据库信息科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019029736A1 (en) * | 2017-08-10 | 2019-02-14 | 华为技术有限公司 | Time-domain stereo coding and decoding method and related product |
| US11355131B2 (en) | 2017-08-10 | 2022-06-07 | Huawei Technologies Co., Ltd. | Time-domain stereo encoding and decoding method and related product |
| US11900952B2 (en) | 2017-08-10 | 2024-02-13 | Huawei Technologies Co., Ltd. | Time-domain stereo encoding and decoding method and related product |
| WO2021000724A1 (en) * | 2019-06-29 | 2021-01-07 | 华为技术有限公司 | Stereo coding method and device, and stereo decoding method and device |
| US11887607B2 (en) | 2019-06-29 | 2024-01-30 | Huawei Technologies Co., Ltd. | Stereo encoding method and apparatus, and stereo decoding method and apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104347077B (en) | 2018-01-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101933086B (en) | Method and apparatus for processing audio signal | |
| CN101202043B (en) | Method and system for encoding and decoding audio signal | |
| CN100395817C (en) | Encoding device and decoding device | |
| CN1973320B (en) | Stereo coding and decoding methods and apparatuses thereof | |
| CN1151705C (en) | Method and apparatus for encoding and decoding multiple audio channels at low bit rates | |
| NZ599981A (en) | Decoding of multichannel audio encoded bit streams using adaptive hybrid transformation | |
| CN100571043C (en) | A kind of space parameter stereo coding/decoding method and device thereof | |
| CN101556799B (en) | Audio decoding method and audio decoder | |
| CN105103225A (en) | Stereo audio encoder and decoder | |
| CN100561576C (en) | A kind of based on the stereo of quantized singal threshold and multichannel decoding method and system | |
| CA2717584A1 (en) | Method and apparatus for processing an audio signal | |
| CN104471641A (en) | Method and device for improving the rendering of multi-channel audio signals | |
| CN1822508B (en) | Method and apparatus for encoding and decoding digital signals | |
| CN101836250A (en) | The method and the device that are used for processing signals | |
| CN103106902A (en) | Low bit-rate audio signal coding and/or decoding method | |
| CN106373583B (en) | Multi-audio-frequency object coding and decoding method based on ideal soft-threshold mask IRM | |
| CA2603027A1 (en) | Device and method for generating a data stream and for generating a multi-channel representation | |
| KR20050087956A (en) | Lossless audio decoding/encoding method and apparatus | |
| CN103814406B (en) | For transition coding or the analysis of decoding or the self adaptation of synthetic weighted window | |
| CN107430867A (en) | Decode the audio bit stream of the frequency spectrum tape copy metadata at least one filling element with enhancing | |
| CN102982805A (en) | Multi-channel audio signal compressing method based on tensor decomposition | |
| CN105308680A (en) | Audio encoder and decoder | |
| CN107112024A (en) | The coding and decoding of audio signal | |
| TWI631554B (en) | Encoding device and method, decoding device and method, and program | |
| CN110660401B (en) | Audio object coding and decoding method based on high-low frequency domain resolution switching |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |