CN101572088A - Stereo encoding and decoding method, a coder-decoder and encoding and decoding system - Google Patents

Stereo encoding and decoding method, a coder-decoder and encoding and decoding system Download PDF

Info

Publication number
CN101572088A
CN101572088A CNA2008100944366A CN200810094436A CN101572088A CN 101572088 A CN101572088 A CN 101572088A CN A2008100944366 A CNA2008100944366 A CN A2008100944366A CN 200810094436 A CN200810094436 A CN 200810094436A CN 101572088 A CN101572088 A CN 101572088A
Authority
CN
China
Prior art keywords
code
signal
decoding
stereosonic
code check
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.)
Pending
Application number
CNA2008100944366A
Other languages
Chinese (zh)
Inventor
鲍长春
贾懋珅
李锐
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Beijing University of Technology
Original Assignee
Huawei Technologies Co Ltd
Beijing University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd, Beijing University of Technology filed Critical Huawei Technologies Co Ltd
Priority to CNA2008100944366A priority Critical patent/CN101572088A/en
Publication of CN101572088A publication Critical patent/CN101572088A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Compression, Expansion, Code Conversion, And Decoders (AREA)

Abstract

The embodiment of the invention discloses a stereo encoding and decoding method, a coder-decoder and an encoding and decoding system, which belong to the technical field of multimedia and can encode and decode an embedded stereo signal. The stereo encoding method comprises the following steps: judging the sampling rate of a stereo signal and selecting a corresponding encoding mode; calculating the average value and the difference value to a left channel and a right channel of the stereo signal to respectively obtain a median signal and a sideband signal of the stereo signal; and respectively encoding the median signal and the sideband signal to form a code steam with an embedded structure. Relevant to the encoding process, the stereo decoding method comprises the following steps: decoding the code stream with the embedded structure by selecting different code rate layers, respectively calculating a median signal and a sideband signal, and obtaining the stereo signal by the synthesis of the median signal and the sideband signal. The embodiment can process a wideband stereo signal sampled by 16 kHz and an ultra wideband stereo signal sampled by 32 kHz by building a plurality of code rate layers.

Description

Stereo coding/decoding method, codec and coding/decoding system
Technical field
The present invention relates to multimedia technology field, particularly relate to a kind of stereo coding/decoding method, codec and coding/decoding system.
Background technology
Be accompanied by the fast development of mobile multimedia technology, voice, audio coding technology have obtained application more and more widely, and along with the increase of network transmission capacity, people also have higher requirement for network institute signal transmission quality simultaneously.Because stereophonic signal can provide good direction feeling, gives sensation on the spot in person, therefore, be applicable to that voice, the audio stereo signal of Network Transmission is encoded into a hot issue of studying at present.
For stereo coding, popular audio coding standard has at present: third generation partner program (3rdGeneration Partnership Project, 3GPP) enhancing AMR-WB standard (the Enhanced Adaptive Multi-Rate-Wideband of Zhi Dinging, AMR-WB+), and mobile motion picture expert group version (Moving Picture Expert Group, MPEG) serial coding standards such as the MPEG-1 of Zhi Dinging, MPEG-2, MPEG-4.In AMR-WB+ stereo coding standard, stereophonic signal is divided into two parts of high and low frequency, for low frequency part, their left and right acoustic channels of encoding and, difference signal; For HFS, two sound channels are carried out absolute coding.In MPEG series stereo coding standard, what adopt is intensity-stereo encoding, further remove redundance and uncorrelated degree between the sound channel according to the correlativity between two sound channels and binaural hearing effect, the method of removing the degree of correlation is the M/S method, be about to the frequency spectrum addition of 2 sound channels or subtract each other, according to the size decision of the degree of correlation of two sound channels be transmission and, difference signal or transmission original signal.
In realizing process of the present invention, the inventor finds, stereo coding of the prior art, be based on the fixed code stream encryption or encode based on the selection of code check, when limited at the network bandwidth or occur under the situation of code stream packet loss, because decoding end can not receive complete encoding code stream, will cause to decode and signal occurs and interrupt suddenly.
Summary of the invention
The technical matters that the embodiment of the invention will solve provides a kind of stereo coding/decoding method, codec and coding/decoding system, can encoding and decoding embedded stereoscopic acoustical signal, guarantee the continuity of left and right sound track signals.
For achieving the above object, the embodiment of the invention is achieved by the following technical solution:
The embodiment of the invention provides a kind of stereo encoding method, may further comprise the steps:
The sampling rate of stereophonic signal is judged, selects the respective coding pattern;
According to the coding mode of selecting, difference is averaged, asked to the left and right acoustic channels of stereophonic signal, obtains the med signal and the sideband signals of stereophonic signal respectively;
Described med signal and sideband signals are carried out embedded encoded processing respectively, form the code stream of damascene structures.
The embodiment of the invention also provides a kind of stereo encoding apparatus, comprising:
Coding mode selected cell, the sampling rate that is used for stereophonic signal are judged, select the respective coding pattern;
The med signal acquiring unit is used for according to the coding mode of selecting, and the left and right acoustic channels of stereophonic signal is averaged and obtained stereosonic med signal;
The sideband signals acquiring unit is used for according to the coding mode of selecting, and the left and right acoustic channels of stereophonic signal asks difference to obtain stereosonic sideband signals;
The med signal coding unit is used for the med signal of described stereophonic signal is carried out embedded encoded, obtains the code stream of damascene structures;
The sideband signals coding unit is used for sideband signals to described stereophonic signal and carries out embedded encodedly, obtains the code stream of damascene structures.
The embodiment of the invention also provides a kind of stereo decoding method, may further comprise the steps:
Sampling rate to the code stream of damascene structures is judged, selects corresponding decoding schema;
According to the decoding schema of selecting, the code rate layer that the code stream selection of damascene structures is different is decoded, solve med signal and sideband signals respectively;
Med signal that decoding is obtained and sideband signals are sued for peace and are asked poor, try to achieve stereosonic left and right acoustic channels respectively, the synthetic stereophonic signal that obtains.
The embodiment of the invention also provides a kind of stereo decoding apparatus, comprising:
The decoding schema selected cell is used for the sampling rate of the code stream of damascene structures is judged, selects corresponding decoding schema;
The code rate layer selected cell is used for according to the decoding schema of selecting, the code rate layer different to the code stream selection of damascene structures;
The med signal decoding unit is used for the part code rate layer decoding of selecting is obtained stereosonic med signal;
The sideband signals decoding unit is used for the part code rate layer decoding of selecting is obtained stereosonic sideband signals;
The stereophonic signal synthesis unit is used for med signal that decoding is obtained and sideband signals and sues for peace and ask poor, tries to achieve stereosonic left and right acoustic channels respectively, the synthetic stereophonic signal that obtains.
The embodiment of the invention also provides a kind of stereo coding system, comprising:
Coding mode selected cell, the sampling rate that is used for stereophonic signal are judged, select the respective coding pattern;
The broadband stereo encoding apparatus is used for difference is averaged, asked to the stereosonic left and right acoustic channels in broadband, obtains the med signal and the sideband signals of broadband stereophonic signal respectively; Described med signal and sideband signals are carried out embedded encoded processing respectively, form the code stream of broadband damascene structures;
The ultra broadband stereo encoding apparatus is used for difference is averaged, asked to the stereosonic left and right acoustic channels of ultra broadband, obtains the med signal and the sideband signals of ultra broadband stereophonic signal respectively; Described med signal and sideband signals are carried out embedded encoded processing respectively, form the code stream of ultra broadband damascene structures.
The embodiment of the invention also provides a kind of stereo decoding system, comprising:
The decoding schema selected cell is used for the sampling rate of the code stream of damascene structures is judged, selects corresponding decoding schema;
The broadband stereo decoding apparatus, be used for the different code rate layer of the code stream selection of the stereosonic damascene structures in broadband is carried out embedded decoding, obtain stereosonic med signal in broadband and sideband signals respectively, suing for peace and asking to differ from obtains stereosonic left and right acoustic channels, the synthetic broadband stereophonic signal that obtains;
The ultra broadband stereo encoding apparatus, be used for the different code rate layer of the code stream selection of the stereosonic damascene structures of ultra broadband is carried out embedded decoding, obtain stereosonic med signal of ultra broadband and sideband signals respectively, suing for peace and asking to differ from obtains stereosonic left and right acoustic channels, the synthetic ultra broadband stereophonic signal that obtains.
By above technical scheme as can be known, at coding side, difference is averaged, asked to left and right acoustic channels by stereophonic signal, obtain the med signal and the sideband signals of stereophonic signal respectively, then described med signal and sideband signals are carried out embedded encoded processing respectively, form the code stream of damascene structures, realize the embedded encoded of stereophonic signal; In decoding end, by the different code rate layer of the code stream selection of damascene structures is carried out embedded decoding, solve med signal and sideband signals respectively, then described med signal and sideband signals are sued for peace and ask poor, try to achieve stereosonic left and right acoustic channels, the synthetic stereophonic signal that obtains is realized the decoding to stereosonic damascene structures code stream.The embodiment of the invention is by carrying out the code stream of embedded encoded formation damascene structures in the coding side stereophonic signal, carry out corresponding embedded decoding in decoding end by code stream to damascene structures, do not having under the situation of high-rise code stream information still can be separately the code stream of low layer to be carried out program, guarantee the continuity of stereo left and right sound track signals, strengthened the robustness of stereophonic signal when Channel Transmission.
Description of drawings
The process flow diagram of the stereo encoding method that Fig. 1 provides for the embodiment of the invention one;
The process flow diagram of the broadband stereo encoding method that Fig. 2 provides for the embodiment of the invention two;
The MLT encoder encodes theory diagram that Fig. 3 provides for the embodiment of the invention;
The code flow structure figure of the MLT scrambler that Fig. 4 provides for the embodiment of the invention;
The code stream hierarchical diagram of the MLT scrambler that Fig. 5 provides for the embodiment of the invention;
The process flow diagram of the ultra broadband stereo encoding method that Fig. 6 provides for the embodiment of the invention three;
The structural drawing of the stereo encoding apparatus that Fig. 7 provides for the embodiment of the invention four;
A kind of 16kHz sampling rate broadband stereo encoding apparatus structural drawing that Fig. 8 provides for the embodiment of the invention five;
A kind of ultra broadband stereo encoding apparatus structural drawing that Fig. 9 provides for the embodiment of the invention six;
The 32kHz sampling rate ultra broadband stereo encoding apparatus structural drawing that Figure 10 provides for the embodiment of the invention seven;
The process flow diagram of the stereo decoding method that Figure 11 provides for the embodiment of the invention eight;
The MLT territory coding/decoding method theory diagram that Figure 12 provides for the embodiment of the invention;
The structural drawing of the stereo decoding apparatus that Figure 13 provides for the embodiment of the invention nine;
The structural drawing of the 16kHz sampling rate broadband stereo decoding apparatus that Figure 14 provides for the embodiment of the invention ten;
The structural drawing of a kind of ultra broadband stereo decoding apparatus that Figure 15 provides for the embodiment of the invention 11;
The structural drawing of the ultra broadband stereo decoding apparatus of the 32kHz sampling rate that Figure 16 provides for the embodiment of the invention 12;
The structural drawing of the stereo decoding system that stereo coding system that Figure 17 provides for the embodiment of the invention 13 and embodiment 14 provide.
Embodiment
The technical scheme of the embodiment of the invention be based on the embedded type speech encoding standard G.729.1 and modulated lapped transform (mlt) (Modulated Lapped Transform, MLT) encryption algorithm generate a cover stereophonic encoder.This scrambler can either coding and decoding 16kHz sampling rate the broadband stereophonic signal, can handle the ultra broadband stereophonic signal of 32kHz sampling again.Below in conjunction with Figure of description the embodiment of the invention is elaborated.
Referring to Fig. 1, the process flow diagram of the stereo encoding method that provides for the embodiment of the invention one comprises step:
Step 101, the sampling rate of stereophonic signal are judged, select the respective coding pattern;
If judge that the sampling rate of stereophonic signal is 16kHz, then select broadband stereo coding pattern; If judge that the sampling rate of stereophonic signal is 32kHz, then select ultra broadband stereo coding pattern.
For the broadband stereophonic signal of coding 16kHz sampling, signal effective bandwidth is 7kHz; The ultra broadband stereophonic signal of coding 32kHz sampling, the effective bandwidth of processing signals is 14kHz.The code stream that total encoder encodes forms is Embedded, and the stream packet of low layer code check is contained in the code stream of high layer bit rate, and the decoded signal quality is successively increased progressively to high level by low layer.
Step 102, according to the coding mode of selecting, difference is averaged, asked to the left and right acoustic channels of stereophonic signal, obtains the med signal and the sideband signals of stereophonic signal respectively;
Step 103 is carried out embedded encoded processing respectively to described med signal and sideband signals, forms the code stream of damascene structures.
Below and processing procedure ultra broadband stereo coding pattern stereo to the broadband describe respectively:
As shown in Figure 2, the process flow diagram of the broadband stereo encoding method that provides for the embodiment of the invention two comprises step:
Difference is averaged, asked to step 201 to the left and right acoustic channels of broadband stereophonic signal, obtains the med signal and the sideband signals of broadband stereophonic signal respectively;
From the stereophonic signal of input broadband, extract left and right sound track signals signal_wb respectively Left(n) and signal_wb Right(n), the difference operation is averaged, asked to left and right sound track signals according to formula (1), obtain the med signal signal_wb of broadband stereophonic signal respectively Mid(n) and sideband signals signal_wb Side(n).
signal _ wb mid ( n ) = signal _ wb left ( n ) + signal _ wb right ( n ) 2 (1)
signal _ wb side ( n ) = signal _ wb left ( n ) - signal _ wb right ( n ) 2
Wherein, med signal is the coherent signal of stereo left and right sound channels, and sideband signals has been represented the difference of two sound channel signals, and these two signals all are the broadband signals of 16kHz sampling.
Step 202, it is the embedded encoded of 32kbps that described med signal is carried out the all-key rate, obtains the code stream between 0~32kbps code check;
To described med signal signal_wb Mid(n) can by adopt the embedded type speech encoding standard G.729.1 coder processes obtain code stream between 0~32kbps code check, but be not limited to G.729.1 scrambler, the embedded encoder that to go for any all-key rate be 32kbps.
Step 203, it is embedded encoded that described sideband signals is carried out the 16kbps code check, obtains the code stream between 32kbps~48kbps code check.
To described sideband signals signal_wb Side(n) can carry out 16kbps code check coding by adopting MLT, obtain the code stream between 32kbps~48kbps code check, can certainly have the scrambler of said function by adopting other.
MLT encoder encodes principle is introduced the MLT coding method as shown in Figure 3 below in detail.
At first, the 16kHz sideband signals to input carries out the MLT conversion according to formula (2):
mlt ( m ) = Σ n = 0 639 2 320 sin ( π 640 ( n + 0.5 ) ) cos ( π 320 ( n - 159.5 ) ( m + 0.5 ) ) x ( n ) - - - ( 2 )
Obtain 320 MLT coefficients, these value representations be spectrum value in frequency 0~8kHz scope, because the bandwidth range that broadband signal requires to handle is 0~7kHz, therefore extract the MLT coefficient of 0~7kHz frequency band range and encode, promptly extract preceding 280 MLT coefficients and handle.
The MLT coefficient that extracts is divided into 14 subbands equably, is also referred to as 14 zones, each subband has 20 sampling points, and the frequency span of expression is 500Hz.
Obtain the average amplitude envelope value of each subband according to formula (3), the amplitude envelops value of subband r is defined as the root mean square of MLT coefficient in this zone.
rms ( r ) = 1 20 Σ n = 0 19 mlt ( 20 r + n ) mlt ( 20 r + n ) - - - ( 3 )
Amplitude envelops value to 14 zones obtaining is carried out scalar quantization, and the quantization index value is carried out writing code stream after the Huffman encoding.
Size according to each subband amplitude envelops quantitative information, line ordering brought in antithetical phrase, stipulates the importance of each subband simultaneously, and the subband that the amplitude envelops value is big is attached most importance to and wanted subband, important in the inferior important subband of the little subband of envelope value, this operation is used for the later stage and writes code stream.
According to amplitude envelops quantitative information that obtains previously and the available bit number of coding MLT coefficient, can obtain 16 kinds of signal sorting techniques, the bit number that each subband distributed under the different mode classifications is different.From 16 kinds of sorting techniques, select the optimal classification method, describe, and these 4 classification control bits are write code stream with the classification control information of 4 bits.
MLT coefficient in each subband all divided by this subband amplitude envelops quantized value, is obtained the MLT coefficient after the normalization.To the coefficient after these normalization, carry out scalar quantization, according to the optimal classification method of selecting, 20 index values that quantize to obtain are divided into several groups of vectors, and these vectors are carried out Huffman encoding.
After the MLT coefficient is carried out Huffman encoding, write code stream according to the importance ranking of each subband of trying to achieve previously, the MLT quantization index in important zone writes code stream earlier, writes code stream behind the index in inferior important zone.
The code stream information that final coding forms is made of three parts as shown in Figure 4: the amplitude envelops quantitative information (elongated) of each subband, classification control information (4 bit) and MLT coefficient quantization index information (elongated).
Because after adding this part MLT transition coding, the total bitrate of scrambler expands to 40kbps and 48kbps, so every frame bit number of 40kbps coding layer MLT coefficient is: (40-32) kbps*20ms=160 bit; Every frame bit number of 48kbps coding layer MLT coefficient is: (48-40) kbps*20ms=160 bit.
When coding, what carry out all is the coding of MLT coefficient under the 16kbps code check of 7kHz bandwidth range, forms the code stream information of every frame 320 bits.If coding side is indicated, with the code check coding of 40kbps, then requiring the code check to the MLT coefficient coding is 8kbps, and therefore need block the code stream of 320 bit informations of formation, preceding 160 bit informations that keep 320 bit informations are used for transmission, abandon remaining bit information.Because when carrying out Huffman encoding, be to sort from high to low according to sub belt energy to write code stream successively, the amplitude envelops quantitative information of each subband, classification control information simultaneously all write in preceding 80 bit informations, and therefore preceding 160 bit information is important in far away back 160 bits in 320 bit informations.
Fig. 5 is concrete code stream hierarchical diagram, as shown in Figure 5, the 40kbps layer bit stream has comprised the MLT coefficient index information of amplitude envelope quantized value, classification control information and the important subband of each frequency band, and the 48kbps layer bit stream has included only the MLT coefficient index information of time important subband.
As shown in Figure 6, the process flow diagram of the ultra broadband stereo encoding method that provides for the embodiment of the invention three comprises step:
Difference is averaged, asked to step 301 to the left and right acoustic channels of ultra broadband stereophonic signal, obtains the med signal and the sideband signals of ultra broadband stereophonic signal respectively;
When the input be the ultra broadband stereophonic signal of 32kHz sampling rate the time, at first, from input signal, extract left and right sound track signals signal_swb respectively Left(n) and signal_swb Right(n).The difference operation is averaged, asked to left and right sound track signals respectively, obtain the med signal signal_swb of input signal by formula (4) Mid(n) and sideband signals signal_swb Side(n):
signal _ swb mid ( n ) = signal _ swb left ( n ) + signal _ swb right ( n ) 2 (4)
signal _ swb side ( n ) = signal _ swb left ( n ) - signal _ swb right ( n ) 2
Wherein, med signal signal_swb Mid(n) be the coherent signal of stereo left and right sound channels, sideband signals signal_swb Side(n) represented the difference of two sound channel signals, these two signals all are the ultra-broadband signals of 32kHz sampling.
Step 302, it is the embedded encoded of 32kbps that the information of described med signal 0~7kHz frequency band is carried out the all-key rate, obtains the code stream between 0~32kbps code check;
To the stereosonic med signal signal_swb of the ultra broadband of 32kHz sampling rate Mid(n) descend 2 broadband signals that sample 16kHz sampling, to this signal can by adopt the embedded type speech encoding standard G.729.1 coder processes obtain code stream between 0~32kbps code check; But be not limited to G.729.1 scrambler, the embedded encoder that to go for any all-key rate be 32kbps.
Step 303 is carried out the embedded encoded of 16kbps code check to the information of the 7kHz~14kHz frequency band of described med signal, obtains the code stream between 32kbps~48kbps code check;
Med signal signal_swb to ultra broadband Mid(n) can adopt MLT to carry out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, can certainly adopt other to have the scrambler of said function.Below be to adopt MLT transition coding process:
With med signal signal_swb Mid(n) carry out the MLT conversion according to formula (5):
mlt ( m ) = Σ n = 0 1279 2 640 sin ( π 1280 ( n + 0.5 ) ) cos ( π 640 ( n - 319 . 5 ) ( m + 0.5 ) ) x ( n ) - - - ( 5 )
Obtain 640 MLT coefficients, these value representations be spectrum value in frequency 0~16kHz scope, extract the MLT coefficient of 7~14kHz frequency band range and encode, obtain 280 MLT coefficients.
To these 280 MLT coefficient codings, employing be and the stereo 40kbps in broadband, MLT transform domain coding mode that the 48kbps code rate layer is identical.The MLT coefficient with the coding that the 16kbps code check carries out, is formed the code stream information of every frame 320 bits.What write in the code stream is the amplitude envelops quantitative information of the MLT coefficient of each subband, the MLT coefficient quantization index information of classification control bit information and each subband, and wherein MLT coefficient quantization index information is that importance according to subband writes code stream from high to low successively.If coding side indicates that with the code check coding of 40kbps, promptly requiring the code check to the MLT coefficient coding is 8kbps, preceding 160 bit informations that then keep 320 bit informations are used for transmission, abandon remaining bit information.
Step 304 is carried out the embedded encoded of 8kbps code check to the information of 0~7kHz frequency band of described sideband signals, obtains the code stream between 48kbps~56kbps code check;
Step 305 is carried out the embedded encoded of 8kbps code check to the information of described sideband signals 7kHz~14kHz frequency band, obtains the code stream between 56kbps~64kbps code check.
Step 304 and step 305 are to sideband signals signal_swb Side(n) processing of carrying out in 56kbps, 64kbps code rate layer, can carry out 8kbps code check coding by adopting MLT, wherein 56kbps code rate layer coding is the MLT coefficient of sideband signals in 0~7kHz spectral range, and what the 64kbps code rate layer was encoded is the MLT coefficient of sideband signals in 7~14kHz spectral range.This two-layer when coding adopts all is and 40kbps, MLT transform domain coding mode that the 48kbps code rate layer is identical, and difference is, at this two-layer be to bandwidth be the MLT coefficient of 7kHz with the coding that the 8kbps code check carries out, form the code stream information of every frame 160 bits.Like this, the code stream of 56kbps code rate layer has comprised the amplitude envelope quantitative information of each subband in sideband signals 0~7kHz scope, the MLT coefficient quantization index information of classify control information, all subbands; The code stream of 64kbps code rate layer has comprised the MLT coefficient quantization index information of amplitude envelope quantitative information, classification control information and all subbands of each subband in sideband signals 7~14kHz scope.
On the basis of the stereo encoding method that the embodiment of the invention provides, the embodiment of the invention four provides a kind of stereo encoding apparatus, as shown in Figure 7, comprising:
Coding mode selected cell 600, the sampling rate that is used for stereophonic signal are judged, select the respective coding pattern;
Med signal acquiring unit 601 is used for according to the coding mode of selecting, and the left and right acoustic channels of stereophonic signal is averaged and obtained the med signal of stereophonic signal;
Sideband signals acquiring unit 602 is used for according to the coding mode of selecting, and the left and right acoustic channels of stereophonic signal asks difference to obtain the sideband signals of stereophonic signal;
Med signal coding unit 603 is used for the med signal of described stereophonic signal is carried out embedded encoded, forms the code stream of damascene structures;
Sideband signals coding unit 604 is used for sideband signals to described stereophonic signal and carries out embedded encodedly, forms the code stream of damascene structures.
When stereo when being the broadband stereophonic signal, described med signal coding unit 603, being used for the stereosonic med signal in broadband is carried out the all-key rate is the embedded encoded of 32kbps, forms the code stream between 0~32kbps code check; Described sideband signals coding unit 604 is used for the stereosonic sideband signals in broadband is carried out the embedded encoded of 16kbps code check, forms the code stream between 32kbps~48kbps code check.
As shown in Figure 8, be a kind of broadband stereo encoding apparatus that the embodiment of the invention five provides, the broadband stereophonic signal for the 16kHz sampling rate comprises:
Med signal acquiring unit 11 and sideband signals acquiring unit 12, described med signal acquiring unit 11 and sideband signals acquiring unit 12 obtain stereosonic med signal in broadband and sideband signals respectively according to formula (1).
G.729.1 scrambler 13, be used for by adopt the embedded type speech encoding standard G.729.1 scrambler the stereosonic med signal in broadband is carried out the all-key rate is the embedded encoded of 32kbps, obtain the code stream between 0~32kbps code check;
40kbps layer coding module 14, be used for by adopting the modulated lapped transform (mlt) scrambler that the stereosonic sideband signals in broadband is carried out the embedded encoded of 16kbps code check, obtain the code stream between 32kbps~48kbps code check, and comprised the modulated lapped transform (mlt) coefficient quantization index information of the amplitude envelops quantitative information of each subband, classification control information and important subband in the sideband signals in the code stream;
48kbps layer coding module 15, be used for by adopting the modulated lapped transform (mlt) scrambler that the stereosonic sideband signals in broadband is carried out the embedded encoded of 16kbps code check, obtain the code stream between 32kbps~48kbps code check, and included only the modulated lapped transform (mlt) coefficient quantization index information of time important subband in the sideband signals in the code stream.
When selecting ultra broadband stereo coding pattern, as shown in Figure 9, a kind of ultra broadband stereo encoding apparatus for the embodiment of the invention six provides comprises:
Med signal acquiring unit 601 is used for the left and right acoustic channels of ultra broadband stereophonic signal averaged and obtains the stereosonic med signal of ultra broadband;
Sideband signals acquiring unit 602 is used for asking difference to obtain the stereosonic sideband signals of ultra broadband to the left and right acoustic channels of ultra broadband stereophonic signal;
Med signal coding unit 603, being used for the stereosonic med signal of described ultra broadband is carried out the all-key rate is the embedded encoded of 48kbps, forms the code stream between 0~48kbps code check;
Sideband signals coding unit 604 is used for the stereosonic sideband signals of described ultra broadband is carried out the embedded encoded of 16kbps code check, forms the code stream between 48kbps~64kbps code check.
Wherein, described med signal coding unit 603 specifically comprises:
Med signal low frequency coding module 603A is used for the low frequency part to the stereosonic med signal of ultra broadband, and carrying out the all-key rate is the embedded encoded of 32kbps, obtains the code stream between 0~32kbps code check;
Med signal high-frequency coding module 603B is used for the HFS to the stereosonic med signal of ultra broadband, carries out the embedded encoded of 16kbps code check, obtains the code stream between 32kbps~48kbps code check.
Described sideband signals coding unit 604 specifically comprises:
Sideband signals low frequency coding module 604A is used for the low frequency part of the stereosonic sideband signals of ultra broadband is carried out the embedded encoded of 8kbps code check, obtains the code stream between 48kbps~56kbps code check;
Sideband signals high-frequency coding module 604B is used for the HFS of the stereosonic sideband signals of ultra broadband is carried out the embedded encoded of 8kbps code check, obtains the code stream between 56kbps~64kbps code check.
As shown in figure 10, be a kind of concrete ultra broadband stereo encoding apparatus that the embodiment of the invention seven provides, the ultra broadband stereophonic signal for the 32kHz sampling comprises:
Med signal acquiring unit 21 and sideband signals acquiring unit 22, described med signal acquiring unit 21 and sideband signals acquiring unit 22 obtain stereosonic med signal of ultra broadband and sideband signals respectively according to formula (4).
G.729.1 scrambler 23, corresponding med signal low frequency coding module 603A, by adopting G.729.1 scrambler of embedded type speech encoding standard, be used for obtaining the code stream between 0~32kbps code check to 2 sampling the coding that 16kHz carries out the 32kbps code check under the stereosonic med signal of ultra broadband;
40kbps layer coding module 24, corresponding med signal high-frequency coding module 603B, be used for by adopting modulated lapped transform (mlt) that the med signal of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, comprised the amplitude envelops quantitative information of interior each subband of 7kHz~14kHz frequency band range of med signal, the modulated lapped transform (mlt) coefficient quantization index information of classify control information and important subband in the described code stream;
48kbps layer coding module 25, corresponding med signal high-frequency coding module 603B, be used for by adopting modulated lapped transform (mlt) that the med signal of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, included only the modulated lapped transform (mlt) coefficient quantization index information of the interior time important subband of 7kHz~14kHz frequency band range of med signal in the described code stream;
56kbps layer coding module 26, corresponding described sideband signals low frequency coding module 604A, be used for by adopting modulated lapped transform (mlt) that the transform domain information of the stereosonic sideband signals 0~7kHz of ultra broadband frequency band part is carried out the embedded encoded of 8kbps code check, obtain the code stream between 48kbps~56kbps code check, comprised the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and all subbands of each subband in sideband signals 0~7kHz frequency band range in the described code stream;
64kbps layer coding module 27, corresponding described sideband signals high-frequency coding module 604B, be used for by adopting modulated lapped transform (mlt) that the transform domain information of the stereosonic sideband signals 7kHz of ultra broadband~14kHz frequency band part is carried out the embedded encoded of 8kbps code check, obtain the code stream between 56kbps~64kbps code check, comprised the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and all subbands of each subband in sideband signals 7kHz~14kHz frequency band range in the described code stream.
Corresponding with stereo encoding method, as shown in figure 11, the process flow diagram of the stereo decoding method that provides for the embodiment of the invention eight comprises the steps:
Step 401 is judged the sampling rate of the code stream of damascene structures, selects corresponding decoding schema;
If judge that the sampling rate of the code stream of damascene structures is 16KHz, then select the stereosonic decoding schema in broadband; If judge that the sampling rate of the code stream of damascene structures is 32KHz, then select the stereosonic decoding schema of ultra broadband.
Step 402 according to the decoding schema of selecting, is decoded to the code rate layer that the code stream selection of damascene structures is different, solves med signal and sideband signals respectively;
Step 403, med signal that decoding is obtained and sideband signals are sued for peace and are asked poor, try to achieve stereosonic left and right acoustic channels, the synthetic stereophonic signal that obtains.
Corresponding with coding side, the decoding process to the code stream of the damascene structures of broadband and ultra broadband stereophonic signal describes respectively below.
For the code stream of broadband stereophonic signal, to the code stream between 0~32kbps code check, carrying out the all-key rate is the embedded decoding of 32kbps, obtains the stereosonic med signal in broadband; Code stream to 32kbps~48kbps code check part carries out the decoding of 16kbps code check, obtains the stereosonic sideband signals in broadband.
When the code stream information of the broadband stereophonic signal that receives 16kHz sampling, carry out different decode operations according to the difference that receives bit number.
G.729.1, carrying out the embedded type speech encoding standard when 1, code check is less than 32kbps decodes
When bit number<=640 (during code check<=32kbps), only by adopt the embedded type speech encoding standard G.729.1 demoder decode, decode according to the actual bit number that obtains, generate the broadband med signal of 16kHz sampling under this code check, and sideband signals obtains without any code stream information, therefore sideband signals is put 0.Finally tried to achieve the information of left and right acoustic channels by med signal and sideband signals, obtain the output voice of two-channel, by formula calculate (6):
signal_wb left(n)=signal_wb mid(n)+signal_wb side(n)
signal_wb right(n)=signal_wb mid(n)-signal_wb side(n)(6)
Because sideband signals is 0, therefore, be equivalent to med signal is duplicated to left and right acoustic channels, to guarantee that left and right acoustic channels all has signal.
2, the MLT coefficient of 40kbps, 48kbps code rate layer decoding
When bit number=960 (during code check=48kbps), the following code stream information of 32kbps code check utilize the embedded type speech encoding standard G.729.1 demoder decode, generate the broadband intermediate value decoded signal of 16kHz sampling.40kbps, 48kbps layer bit stream information are decoded by adopting MLT territory decoding process as shown in figure 12 simultaneously.
At first recover the quantized value of each subband amplitude envelope in MLT territory, the classification control information of the available bit number 320 of this value and decoding end, the 4bit that receives is combined, draw the coding specification method of coding side by adopting.Decoding end is by adopting the sorting technique identical with coding side, bit information is carried out inverse quantization handle the MLT coefficient that obtains after each subband removes envelope, and the MLT coefficient that each subband solutions gets be multiply by the amplitude envelope quantized value of this subband, finally obtained the MLT coefficient of signal.Decode this moment is MLT coefficient in 0~7kHz scope, and (7~8kHz) MLT coefficient zero setting is carried out the inverse transformation sideband signals of being decoded accordingly to the MLT coefficient then with all the other frequency bands.
With the embedded type speech encoding standard G.729.1 the broadband med signal that goes out of decoder decode combine the broadband stereophonic signal that obtains decoding according to formula (6) with the sideband signals that 32kbps~48kbps decoding obtains.
When bit number=800 (during code check=40kbps), same, the following code stream information of 32kbps code check with the embedded type speech encoding standard G.729.1 demoder decode.Because 40kbps code rate layer code stream information has comprised the MLT coefficient index information of amplitude envelope quantized value, classification control information and the important subband of each subband, therefore only can decode the MLT coefficient of part subband.Recover the quantized value of each subband amplitude envelope in MLT territory during decoding, and it is sorted, obtain the position of important frequency range, those included zones in this 40kbps layer bit stream information that hence one can see that.In conjunction with the classification control information of the 4bit that receives, MLT coefficient code stream information decodes the MLT coefficient of these important subbands.For in 0~7kHz frequency range important subband, have only their amplitude envelops quantitative information in the code stream of 40kbps code rate layer, combining with these amplitude envelops quantitative informations with random number generates the MLT coefficient of these subbands roughly.Behind the final like this MLT coefficient that obtains all subbands, MLT coefficient zero setting with 7~8kHz frequency band, carry out the sideband signals that inverse transformation obtains decoding, with the embedded type speech encoding standard G.729.1 the broadband med signal of demoder output combine the broadband stereophonic signal that obtains decoding according to formula (6).
Hence one can see that, and for the stereo code stream information in broadband, when code check was 48kbps, complete decoding went out stereosonic med signal in broadband and sideband signals, the output of synthetic wideband stereophonic signal; When code check is 40kbps, decode stereosonic med signal in broadband and the stereosonic sideband signals in broadband that contains important sub-band information, the output of synthetic wideband stereophonic signal; When code check smaller or equal to 32kbps, only can decode the stereosonic med signal in broadband, med signal is composed to left and right acoustic channels, as the stereo output in broadband.Therefore guaranteed the continuity of left and right sound track signals, strengthened the robustness when stereophonic signal transmits in channel.
Decode operation for the code stream of ultra broadband stereophonic signal is judged based on code check equally, corresponding cataloged procedure, to the code stream between 0~32kbps code check, carrying out the all-key rate is the embedded decoding of 32kbps, obtains the information of the stereosonic med signal 0~7kHz of ultra broadband frequency band; Code stream to 32kbps~48kbps code check part carries out the decoding of 16kbps code check, obtains the information of the stereosonic med signal 7kHz of ultra broadband~14kHz frequency band; Code stream to 48kbps~56kbps code check part carries out the decoding of 8kbps code check, obtains the information of the stereosonic sideband signals 0~7kHz of ultra broadband frequency band; Code stream to 56kbps~64kbps code check part carries out the decoding of 8kbps code check, obtains the information of the stereosonic sideband signals 7kHz of ultra broadband~14kHz frequency band.
When the code stream information of the ultra broadband stereophonic signal that receives 32kHz sampling, carry out different decode operations according to the difference that receives bit number.
1, code check is smaller or equal to the embedded type speech encoding standard of 32kbps decode operation G.729.1
When bit number<=640 (during code check<=32kbps), by adopting G.729.1 demoder of embedded type speech encoding standard, decode according to the actual bit number that obtains, generate the med signal of the 16kHz sampling under the phase code rate, last 2 samplings obtain the low-frequency information of the ultra broadband med signal of 32kHz sampling, because do not receive sideband signals, can't decode to sideband signals, so this med signal can only be copied to left and right sound channels respectively, obtaining effective bandwidth is the ultra broadband stereophonic signal of 0~7kHz.Gained stereophonic signal left and right sound channels signal is identical in the case.
2, the ultra broadband MLT coefficient decode operation of 40kbps, 48kbps code rate layer
When bit number=960 (during code check=48kbps), the following code stream information of 32kbps code check utilize the embedded type speech encoding standard G.729.1 demoder decode and generate the broadband intermediate value decoded signal of 16kHz sampling.Simultaneously the 40kbps that obtains, 48kbps layer bit stream information are decoded by adopting with broadband 40kbps, transition coding mode that the 48kbps decoding layer is identical, at first recover the quantized value of each subband amplitude envelope in MLT territory, draw the coding specification method according to the amplitude envelops quantitative information.Bit information is carried out inverse quantization handle the MLT coefficient obtain after each subband removes envelope, and the MLT coefficient that each subband solutions gets be multiply by the amplitude envelope quantized value of this frequency band, finally obtained the MLT coefficient of signal.At coding side coding is MLT coefficient in 7~14kHz scope, therefore need the MLT coefficient that decodes is moved the scope to 7~14kHz, and with all the other frequency bands (0~7kHz, 14~16kHz) MLT coefficient zero setting are carried out inverse transformation to it then and are obtained corresponding time-domain signal.With itself and last 2 the sampling after the embedded type speech encoding standard G.729.1 the decoder decode signal plus obtain the ultra broadband med signal.Therefore can't solve sideband signals because do not have 56kbps, 64kbps layer information in the code stream this moment, so this med signal is copied to left and right sound channels respectively, obtains the ultra broadband stereophonic signal under the 48kbps code check.The left and right sound channels signal of gained stereophonic signal is identical in the case.
(during code check=40kbps), same, the following code stream information of 32kbps code check is utilized G.729.1 demoder of embedded type speech encoding standard, and decoding generates broadband intermediate value decoded signal when bit number=800.Because what 40kbps code rate layer code stream information comprised is the MLT coefficient of important subband in 7~14kHz frequency range, can recover the MLT coefficient of these subbands during decoding processing; For in 7~14kHz frequency range important subband, have only their amplitude envelops quantitative information in the code stream of 40kbps code rate layer, combine with these amplitude envelops quantitative informations by adopting the noise filling technique to generate the MLT coefficient of these subbands roughly with random number.Behind the final like this MLT coefficient that obtains all subbands, it is moved the scope to 7~14kHz, carry out inverse transformation and obtain corresponding time-domain signal.With itself and last 2 the sampling after the embedded type speech encoding standard G.729.1 the decoder decode signal plus obtain the ultra broadband med signal.Equally,, can't solve sideband signals,, obtain the ultra broadband stereophonic signal under the 40kbps code check so this med signal can only be copied to left and right sound channels respectively because there is not the sideband signals code stream information.The left and right sound channels signal of gained stereophonic signal is identical in the case.
3, the sideband signals MLT coefficient decode operation of 56kbps, 64kbps code rate layer
(during code check=56kbps), the following code stream information of 48kbps code check obtains med signal by former layer decoders operations when bit number=1120.What the code stream information of 56kbps code rate layer comprised is the MLT coefficient of sideband signals in 0~7kHz scope.This layer bit stream information is decoded by the mode of employing and 40kbp, MLT transform domain coding that the 48kbps code rate layer is identical, recover the quantized value of each frequency band amplitude envelope of MLT territory successively, classification control information, coding specification type, decode the MLT coefficient thus, carry out anti-MLT conversion then and obtain the sideband signals that decoded effective bandwidth is the 32kHz sampling of 0~7kHz.Same, decoding is obtained med signal and sideband signals, it is poor to sue for peace respectively, ask, and obtains the decoding ultra broadband stereophonic signal of the 32kHz sampling under the 56kbps code check.
(during code check=64kbps), same, the following code stream information of 48kbps code check obtains med signal by former layer decoders operations when bit number=1280.What the code stream information of 56kbps code rate layer comprised is the MLT coefficient of limit signal in 0~7kHz scope; What the code stream information of 64kbps code rate layer was described is the MLT coefficient of sideband signals in 7~14kHz scope.The code stream information of 56kbps, 64kbps code rate layer is decoded by the mode that adopts the MLT transform domain coding, and decoding frequency band range is the MLT coefficient of 0~14kHz, and the MLT coefficient is carried out the sideband signals that inverse transformation obtains decoded 32kHz sampling.Decoding is obtained med signal and sideband signals, and it is poor to sue for peace respectively, ask, and obtains the 32kHz sampling ultra broadband stereophonic signal under the 64kbps code check.
Hence one can see that, and when code check was 64kbps, complete decoding went out stereosonic med signal of ultra broadband and the stereosonic sideband signals of ultra broadband that comprises 0~14kHz band information, synthetic ultra broadband stereophonic signal output; When code check is 56kbps, decode the sideband signals of stereosonic med signal of ultra broadband and the stereosonic 0~7kHz frequency band of ultra broadband, synthetic ultra broadband stereophonic signal output; When code check is 40kbps and 48kbps, only decode the stereosonic med signal of ultra broadband, med signal is composed to left and right acoustic channels, as the stereo output of ultra broadband; When code check during smaller or equal to 32kbps, only decode the med signal of the stereosonic 0~7kHz frequency band of ultra broadband, this med signal is composed to left and right acoustic channels, as the stereo output of ultra broadband.Therefore guaranteed the continuity of left and right sound track signals, strengthened the robustness when stereophonic signal transmits in channel.
On the basis of the stereo decoding method that the embodiment of the invention provides, the embodiment of the invention nine provides a kind of stereo decoding apparatus, as shown in figure 13, comprising:
Decoding schema selected cell 800 is used for the sampling rate of the code stream of damascene structures is judged, selects corresponding decoding schema;
Code rate layer selected cell 801 is used for according to the decoding schema of selecting, the code rate layer different to the code stream selection of damascene structures;
Med signal decoding unit 802 is used for the part code rate layer decoding of selecting is obtained stereosonic med signal;
Sideband signals decoding unit 803 is used for the part code rate layer decoding of selecting is obtained stereosonic sideband signals;
Stereophonic signal synthesis unit 804 is used for med signal that decoding is obtained and sideband signals are sued for peace and asked poor, tries to achieve stereosonic left and right acoustic channels, the synthetic stereophonic signal that obtains.
When selecting the code stream of decoding broadband stereophonic signal according to the decoding schema of input, according to code rate layer selected cell 801, described med signal decoding unit 802, selection is to the code stream between 0~32kbps code check, carrying out the all-key rate is the embedded decoding of 32kbps, obtains the stereosonic med signal in broadband; Described sideband signals decoding unit 803 is selected the code stream between 32kbps~48kbps code check, carries out the decoding of 16kbps code check, obtains the stereosonic sideband signals in broadband.
As shown in figure 14, a kind of broadband stereo decoding apparatus that provides for the embodiment of the invention ten, the code stream for the broadband stereophonic signal of 16kHz sampling rate comprises:
Code rate layer selected cell 31 is used for according to the decoding schema of selecting, the code rate layer different to the code stream selection of the stereosonic damascene structures in broadband;
G.729.1 demoder 32, be used for the code stream between 0~32kbps code check, by adopt the embedded type speech encoding standard G.729.1 demoder carry out the embedded decoding that the all-key rate is 32kbps, obtain the stereosonic med signal in broadband;
40kbps layer decoder module 33, when stream rate is 40kbps, the embedded decoding of carrying out the 16kbps code check according to described 40kbps layer bit stream information obtains comprising the MLT coefficient of important subband, and, the MLT coefficient after filling is carried out the sideband signals that the MLT inverse transformation obtains the broadband stereophonic signal by adopting the noise filling technique that the MLT coefficient of the 48kbps decoding layer that lacks in the code stream is filled;
48kbps layer decoder module 34, when stream rate is 48kbps, the MLT coefficient that the embedded decoding of carrying out the 16kbps code check according to described 40kbps and the two-layer code stream information of 48kbps obtains comprising whole subbands carries out the sideband signals that the MLT inverse transformation obtains the broadband stereophonic signal;
Stereophonic signal synthesis unit 35 is used for med signal that decoding is obtained and sideband signals and sues for peace and ask poor, tries to achieve stereosonic left and right acoustic channels, the synthetic broadband stereophonic signal that obtains.
As shown in figure 15, be a kind of ultra broadband stereo decoding apparatus that the embodiment of the invention 11 provides, the code stream for the ultra broadband stereophonic signal comprises:
Code rate layer selected cell 801 is used for according to the decoding schema of selecting, the code rate layer different to the code stream selection of damascene structures;
Med signal decoding unit 802 is used for the code stream between 0~48kbps code check, and carrying out the all-key rate is the embedded decoding of 48kbps, obtains the stereosonic med signal of ultra broadband;
Sideband signals decoding unit 803 is used for the code stream between 48kbps~64kbps code check is carried out the embedded decoding of 16kbps code check, obtains the stereosonic sideband signals of ultra broadband;
Stereophonic signal synthesis unit 804 is used for stereosonic med signal of ultra broadband that decoding is obtained and sideband signals and sues for peace and ask poor, tries to achieve stereosonic left and right acoustic channels, the synthetic ultra broadband stereophonic signal that obtains.
Wherein, described med signal decoding unit 802 specifically comprises:
Med signal low frequency decoder module 802A is used for the code stream between 0~32kbps code check, and carrying out the all-key rate is the embedded decoding of 32kbps, obtains the stereo med signal 0~7kHz of ultra broadband band information;
Med signal high frequency decoder module 802B is used for the code stream between 32kbps~48kbps code check is carried out the embedded decoding of 16kbps code check, obtains the stereo med signal 7kHz of ultra broadband~14kHz band information.
Described sideband signals decoding unit 803 specifically comprises:
Sideband signals low frequency decoder module 803A is used for the code stream between 48kbps~56kbps code check is carried out the embedded decoding of 8kbps code check, obtains the stereo sideband signals 0~7kHz of ultra broadband band information;
Sideband signals high frequency decoder module 803B is used for the code stream between 56kbps~64kbps code check is carried out the embedded decoding of 8kbps code check, obtains the stereo sideband signals 7kHz of ultra broadband~14kHz band information.
As shown in figure 16, a concrete ultra broadband stereo decoding apparatus that provides for the embodiment of the invention 12, the code stream for the ultra broadband stereophonic signal of 32kHz sampling rate comprises:
Code rate layer selected cell 41 is used for the code rate layer different to the code stream selection of damascene structures;
G.729.1 demoder 42, corresponding med signal low frequency decoder module 802A by adopting G.729.1 demoder of embedded type speech encoding standard, is used for the code stream between 0~32kbps code check is decoded, last 2 sample 32kHz, obtain the med signal of the stereosonic 0~7kHz frequency band of ultra broadband;
40kbps layer decoder module 43, corresponding med signal high frequency decoder module 802B, when stream rate is 40kbps, the embedded decoding of carrying out the 16kbps code check according to described 40kbps layer bit stream information obtains comprising the MLT coefficient of important subband, and, the MLT coefficient after filling is carried out the med signal that the MLT inverse transformation obtains the stereosonic 7kHz-14kHz of ultra broadband by adopting the noise filling technique that the MLT coefficient of the 48kbps decoding layer that lacks in the code stream is filled;
48kbps layer decoder module 44, corresponding med signal high frequency decoder module, when stream rate is 48kbps, the MLT coefficient that the embedded decoding of carrying out the 16kbps code check according to described 40kbps and the two-layer code stream information of 48kbps obtains comprising whole subbands carries out the med signal that the MLT inverse transformation obtains the stereosonic 7kHz of ultra broadband~14kHz frequency band;
56kbps layer decoder module 45, corresponding sides band signal low frequency decoder module 803A, be used for when code check is 56kbps, the MLT coefficient that the embedded decoding of carrying out the 8kbps code check according to 56kbps layer bit stream information obtains comprising whole subbands carries out the sideband signals that the MLT inverse transformation obtains the stereosonic 0~7kHz frequency band of ultra broadband;
64kbps layer decoder module 46, corresponding sides band signal high frequency decoder module 803B, be used for when code check is 64kbps, the MLT coefficient that the embedded decoding of carrying out the 8kbps code check according to 64kbps layer bit stream information obtains comprising whole subbands carries out the sideband signals that the MLT inverse transformation obtains the stereosonic 7kHz of ultra broadband~14kHz frequency band;
Stereophonic signal synthesis unit 47 is used for med signal that decoding is obtained and sideband signals and sues for peace and ask poor, tries to achieve stereosonic left and right acoustic channels respectively, the synthetic ultra broadband stereophonic signal that obtains.
On the basis of the stereo encoding apparatus that the embodiment of the invention provides, as shown in figure 17, the stereo coding system for the embodiment of the invention 13 provides comprises:
Coding mode selected cell 910, the sampling rate that is used for stereophonic signal are judged, select the respective coding pattern;
Broadband stereo encoding apparatus 920 is used for difference is averaged, asked to the stereosonic left and right acoustic channels in broadband, obtains the med signal and the sideband signals of broadband stereophonic signal respectively; Described med signal and sideband signals are carried out embedded encoded processing respectively, form the code stream of broadband damascene structures;
Ultra broadband stereo encoding apparatus 930 is used for difference is averaged, asked to the stereosonic left and right acoustic channels of ultra broadband, obtains the med signal and the sideband signals of ultra broadband stereophonic signal respectively; Described med signal and sideband signals are carried out embedded encoded processing respectively, form the code stream of ultra broadband damascene structures.
When the signal of input was the broadband stereophonic signal of 16kHz sampling rate, the broadband stereo encoding apparatus comprised:
G.729.1 scrambler, by adopt the embedded type speech encoding standard G.729.1 scrambler the stereosonic med signal in described broadband is carried out 32kbps code check coding, obtain the code stream between 0~32kbps code check;
40kbps layer coding module, be used for by adopting modulated lapped transform (mlt) that the stereosonic sideband signals in broadband is carried out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, and comprised the modulated lapped transform (mlt) coefficient quantization index information of the amplitude envelops quantitative information of each subband, classification control information and important subband in the sideband signals in the code stream;
48kbps layer coding module, be used for by adopting modulated lapped transform (mlt) that the stereosonic sideband signals in broadband is carried out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, and included only the modulated lapped transform (mlt) coefficient quantization index information of time important subband in the sideband signals in the code stream.
When the signal of input was the ultra broadband stereophonic signal of 32kHz sampling rate, described ultra broadband stereo encoding apparatus comprised:
G.729.1 scrambler, be used for by adopt the embedded type speech encoding standard G.729.1 scrambler obtain the code stream between 0~32kbps code check to 2 sampling 16kHz to carry out 32kbps embedded encoded under the med signal of the stereosonic 0~7kHz frequency band of ultra broadband;
40kbps layer coding module, be used for by adopting modulated lapped transform (mlt) that the med signal of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, comprised the amplitude envelops quantitative information of interior each subband of 7kHz~14kHz frequency band range of med signal, the modulated lapped transform (mlt) coefficient quantization index information of classify control information and important subband in the described code stream;
48kbps layer coding module, be used for by adopting modulated lapped transform (mlt) that the med signal of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, included only the modulated lapped transform (mlt) coefficient quantization index information of the interior time important subband of 7kHz~14kHz frequency band range of med signal in the described code stream;
56kbps layer coding module, be used for by adopting modulated lapped transform (mlt) that the sideband signals of the stereosonic 0~7kHz frequency band of ultra broadband is carried out 8kbps code check coding, obtain the code stream between 48kbps~56kbps code check, comprised the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and all subbands of each subband in sideband signals 0~7kHz frequency band range in the described code stream;
64kbps layer coding module, be used for by adopting modulated lapped transform (mlt) that the sideband signals of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out 8kbps code check coding, obtain the code stream between 56kbps~64kbps code check, comprised the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and all subbands of each subband in sideband signals 7kHz~14kHz frequency band range in the described code stream.
On the basis of the stereo decoding apparatus that the embodiment of the invention provides, still as shown in figure 17, the stereo decoding system for the embodiment of the invention 14 provides comprises:
Decoding schema selected cell 950 is used for the sampling rate of the code stream of damascene structures is judged, selects corresponding decoding schema;
Broadband stereo decoding apparatus 960, be used for the different code rate layer of the code stream selection of the stereosonic damascene structures in broadband is carried out embedded decoding, obtain stereosonic med signal in broadband and sideband signals respectively, suing for peace and asking to differ from obtains stereosonic left and right acoustic channels, the synthetic broadband stereophonic signal that obtains;
Ultra broadband stereo encoding apparatus 970, be used for the different code rate layer of the code stream selection of the stereosonic damascene structures of ultra broadband is carried out embedded decoding, obtain stereosonic med signal of ultra broadband and sideband signals respectively, suing for peace and asking to differ from obtains stereosonic left and right acoustic channels, the synthetic ultra broadband stereophonic signal that obtains.
When what import is the code stream of the broadband stereophonic signal of 16kHz sampling rate, and described broadband stereo decoding apparatus comprises:
G.729.1 demoder is used for the code stream between 0~32kbps code check is carried out the embedded decoding that the all-key rate is 32kbps, obtains the stereosonic med signal in broadband;
40kbps layer decoder module, be used for when code check is 40kbps, the embedded decoding of carrying out the 16kbps code check according to described 40kbps layer bit stream information obtains comprising the MLT coefficient of important subband, and, the MLT coefficient after filling is carried out the sideband signals that the MLT inverse transformation obtains the broadband stereophonic signal by adopting the noise filling technique that the MLT coefficient of the 48kbps decoding layer that lacks in the code stream is filled;
48kbps layer decoder module, be used for when code check is 48kbps, the MLT coefficient that the embedded decoding of carrying out the 16kbps code check according to described 40kbps and the two-layer code stream information of 48kbps obtains comprising whole subbands carries out the sideband signals that the MLT inverse transformation obtains the broadband stereophonic signal.
When what import is the code stream of the ultra broadband stereophonic signal of 32kHz sampling rate, and described ultra broadband stereo decoding apparatus comprises:
G.729.1 demoder, be used for the code stream between 0~32kbps code check, by adopt the embedded type speech encoding standard G.729.1 demoder carry out the embedded decoding that the all-key rate is 32kbps, last 2 sample 32kHz, obtain the med signal of the stereosonic 0~7kHz frequency band of ultra broadband;
40kbps layer decoder module, be used for when code check is 40kbps, the embedded decoding of carrying out the 16kbps code check according to described 40kbps layer bit stream information obtains comprising the MLT coefficient of important subband, and, the MLT coefficient after filling is carried out the med signal that the MLT inverse transformation obtains the stereosonic 7kHz~14kHz of ultra broadband by adopting the noise filling technique that the MLT coefficient of the 48kbps decoding layer that lacks in the code stream is filled;
48kbps layer decoder module, be used for when code check is 48kbps, the MLT coefficient that the embedded decoding of carrying out the 16kbps code check according to described 40kbps and the two-layer code stream information of 48kbps obtains comprising whole subbands carries out the med signal that the MLT inverse transformation obtains the stereosonic 7kHz of ultra broadband~14kHz frequency band;
56kbps layer decoder module, be used for when code check is 56kbps, the MLT coefficient that the embedded decoding of carrying out the 8kbps code check according to 56kbps layer bit stream information obtains comprising whole subbands carries out the sideband signals that the MLT inverse transformation obtains the stereosonic 0~7kHz frequency band of ultra broadband;
64kbps layer decoder module, be used for when code check is 64kbps, the MLT coefficient that the embedded decoding of carrying out the 8kbps code check according to 64kbps layer bit stream information obtains comprising whole subbands carries out the sideband signals that the MLT inverse transformation obtains the stereosonic 7kHz of ultra broadband~14kHz frequency band.
The technical scheme of the embodiment of the invention is based on existing embedded type speech encoding standard G.729.1 codec and MLT codec, realize that a cover has voice, the audio stereo embedded coding and decoding device of multilayer nest code flow structure, can encoding and decoding broadband and ultra broadband stereophonic signal.Stereo for the broadband, by making up two broadband stereo coding layers under 40kbps, the 48kbps code check, to the left and right acoustic channels of input stereo audio signal sue for peace respectively and do poor, obtain med signal and sideband signals, utilize the embedded type speech encoding standard G.729.1 this two parts information to be encoded with the MLT coded system, solve this two-part information respectively in decoding end, the synthetic broadband stereophonic signal that obtains, the maximal rate that the broadband stereophonic signal is encoded is 48kbps; Stereo for ultra broadband, by making up 40kbps, 48kbps, 56kbps, four ultra broadband stereo coding layers under the 64kbps code check, to the left and right acoustic channels of ultra broadband stereophonic signal of input sue for peace respectively and do poor, obtain med signal and sideband signals, (0~48kbps) encodes respectively to this two parts information with modulated lapped transform (mlt) MLT coded system to utilize ultra broadband monophony encoder section, solve this two-part signal in decoding end, the synthetic ultra broadband stereophonic signal that obtains is 64kbps for the code rate of ultra broadband stereophonic signal maximum.
The encoding scheme that the embodiment of the invention proposes, both can handle 8kHz, the monophonic signal of 16kHz sampling, can handle the broadband stereophonic signal of 16kHz sampling and the ultra broadband stereophonic signal of 32kHz sampling again, the signal type that the scrambler that makes the embodiment of the invention propose can be handled is variation more, also can satisfy simultaneously the requirement of in the multimedia communication multi-channel audio signal being handled, the code stream of its generation has mutually nested relation, can come the code check of control transmission according to quality of channel, make the technical scheme of the embodiment of the invention can satisfy high-quality audio/video meeting like this, teleconference, network service, mobile communication, the network congestion control and the third generation and the 4th third-generation mobile communication are to the widespread demand of embedded speech and audio coding.
By the description of above embodiment, those skilled in the art can be well understood to the present invention and can realize by the mode that software adds essential general hardware platform, can certainly pass through hardware, but the former is better embodiment under a lot of situation.Based on such understanding, the part that the technical scheme of the embodiment of the invention contributes to prior art in essence in other words can embody with the form of software product, this computer software product is stored in the storage medium, comprises that some instructions are used so that a subscriber terminal equipment is carried out the described method of each embodiment of the present invention.Here alleged storage medium, as: ROM/RAM, disk, CD etc.
More than to stereo coding/decoding method provided by the present invention, coding and decoding device and coding/decoding system, be described in detail by specific embodiment, what deserves to be explained is, codec and MLT codec are illustrated stereo embedded coding and decoding principle by adopting G.729.1 among the embodiment, but be not limited to codec concrete among the embodiment, can be by adopting other codec with said function.The explanation of above embodiment just is used for helping to understand method of the present invention and thought thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that all can change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.

Claims (52)

1, a kind of stereo encoding method is characterized in that, may further comprise the steps:
The sampling rate of stereophonic signal is judged, selects the respective coding pattern;
According to the coding mode of selecting, difference is averaged, asked to the left and right acoustic channels of stereophonic signal, obtains the med signal and the sideband signals of stereophonic signal respectively;
Described med signal and sideband signals are carried out embedded encoded processing respectively, form the code stream of damascene structures.
2, stereo encoding method according to claim 1 is characterized in that, when selecting broadband stereo coding pattern, described med signal and sideband signals is carried out embedded encoded processing respectively, and the step that forms the code stream of damascene structures comprises:
It is the embedded encoded of 32kbps that described med signal is carried out the all-key rate, obtains the code stream between 0~32kbps code check;
Described sideband signals is carried out the embedded encoded of 16kbps code check, obtain the code stream between 32kbps~48kbps code check.
3, stereo encoding method according to claim 2, it is characterized in that, broadband stereophonic signal for the 16kHz sampling rate, to described med signal by adopt the embedded type speech encoding standard G.729.1 scrambler to carry out the all-key rate be the embedded encoded of 32kbps, obtain the code stream between 0~32kbps code check.
4, stereo encoding method according to claim 2, it is characterized in that, for the broadband stereophonic signal of 16kHz sampling rate, described sideband signals is carried out the embedded encoded of 16kbps code check by adopting the modulated lapped transform (mlt) scrambler, obtain the code stream between 32kbps~48kbps code check; Code stream between described 32kbp s~48kbps code check constitutes 40kbps code rate layer and 48kbps code rate layer, wherein the 40kbps code rate layer comprises the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and the important subband of each subband in the sideband signals, and the 48kbps code rate layer comprises the modulated lapped transform (mlt) coefficient quantization index information of time important subband in the sideband signals.
5, stereo encoding method according to claim 1 is characterized in that, when selecting ultra broadband stereo coding pattern, described med signal and sideband signals is carried out embedded encoded processing respectively, and the step that forms the code stream of damascene structures comprises:
It is the embedded encoded of 32kbps that the information of described med signal 0~7kHz frequency band is carried out the all-key rate, obtains the code stream between 0~32kbps code check;
It is embedded encoded that the information of the 7kHz~14kHz frequency band of described med signal is carried out the 16kbps code check, obtains the code stream between 32kbps~48kbps code check;
It is embedded encoded that the information of 0~7kHz frequency band of described sideband signals is carried out the 8kbps code check, obtains the code stream between 48kbps~56kbps code check;
It is embedded encoded that the information of described sideband signals 7kHz~14kHz frequency band is carried out the 8kbps code check, obtains the code stream between 56kbps~64kbps code check.
6, stereo encoding method according to claim 5, it is characterized in that, ultra broadband stereophonic signal for the 32kHz sampling rate, to the information of described med signal 0~7kHz frequency band by adopt the embedded type speech encoding standard G.729.1 scrambler to carry out the all-key rate be the embedded encoded of 32kbps, obtain the code stream between 0~32kbps code check.
7, stereo encoding method according to claim 5, it is characterized in that, ultra broadband stereophonic signal for the 32kHz sampling rate, information to the 7kHz~14kHz frequency band of described med signal is carried out the embedded encoded of 16kbps code check by adopting the modulated lapped transform (mlt) scrambler, obtains the code stream between 32kbps~48kbps code check; Code stream between described 32kbps~48kbps code check constitutes 40kbps code rate layer and 48kbps code rate layer, and described 40kbps code rate layer comprises the amplitude envelops quantitative information of interior each subband of 7kHz~14kHz frequency band range of med signal, the modulated lapped transform (mlt) coefficient quantization index information of classify control information and important subband, and described 48kbps code rate layer comprises the modulated lapped transform (mlt) coefficient quantization index information of the interior time important subband of 7kHz~14kHz frequency band range of med signal; Described modulated lapped transform (mlt) coefficient quantization index information is to sort from high to low according to sub belt energy to write code stream successively.
8, stereo encoding method according to claim 5, it is characterized in that, ultra broadband stereophonic signal for the 32kHz sampling rate, information to 0~7kHz frequency band of described sideband signals is carried out the embedded encoded of 8kbps code check by adopting the modulated lapped transform (mlt) scrambler, obtains the code stream between 48kbps~56kbps code check; Code stream between described 48kbps~56kbps code check constitutes the 56kbps code rate layer, comprises the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and all subbands of each subband in sideband signals 0~7kHz frequency band range; Described modulated lapped transform (mlt) coefficient quantization index information is to sort from high to low according to sub belt energy to write code stream successively.
9, stereo encoding method according to claim 5, it is characterized in that, ultra broadband stereophonic signal for the 32kHz sampling rate, information to described sideband signals 7kHz~14kHz frequency band is carried out the embedded encoded of 8kbps code check by adopting the modulated lapped transform (mlt) scrambler, obtains the code stream between 56kbps~64kbps code check; Code stream between described 56kbps~64kbps code check constitutes the 64kbps code rate layer, comprises the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and all subbands of each subband in sideband signals 7kHz~14kHz frequency band range; Described modulated lapped transform (mlt) coefficient quantization index information is to sort from high to low according to sub belt energy to write code stream successively.
10, a kind of stereo encoding apparatus is characterized in that, comprising:
Coding mode selected cell, the sampling rate that is used for stereophonic signal are judged, select the respective coding pattern;
The med signal acquiring unit is used for according to the coding mode of selecting, and the left and right acoustic channels of stereophonic signal is averaged and obtained stereosonic med signal;
The sideband signals acquiring unit is used for according to the coding mode of selecting, and the left and right acoustic channels of stereophonic signal asks difference to obtain stereosonic sideband signals;
The med signal coding unit is used for described stereosonic med signal is carried out embedded encoded, obtains the code stream of damascene structures;
The sideband signals coding unit is used for described stereosonic sideband signals is carried out embedded encoded, obtains the code stream of damascene structures.
11, stereo encoding apparatus according to claim 10 is characterized in that, when selecting broadband stereo coding pattern,
Described med signal coding unit, being further used for the stereosonic med signal in broadband is carried out the all-key rate is the embedded encoded of 32kbps, obtains the code stream between 0~32kbps code check;
Described sideband signals coding unit is further used for the stereosonic sideband signals in broadband is carried out the embedded encoded of 16kbps code check, obtains the code stream between 32kbps~48kbps code check.
12, stereo encoding apparatus according to claim 11 is characterized in that, when being the broadband stereophonic signal of 16kHz sampling rate, described med signal coding unit comprises when stereo:
G.729.1 scrambler, be used for by adopt the embedded type speech encoding standard G.729.1 scrambler the stereosonic med signal in broadband is carried out the all-key rate is the embedded encoded of 32kbps, obtain the code stream between 0~32kbps code check.
13, stereo encoding apparatus according to claim 11 is characterized in that, when being the broadband stereophonic signal of 16kHz sampling rate, described sideband signals coding unit comprises when stereo:
40kbps layer coding module, be used for by adopting the modulated lapped transform (mlt) scrambler that the stereosonic sideband signals in broadband is carried out the embedded encoded of 16kbps code check, obtain the code stream between 32kbps~48kbps code check, comprise the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and the important subband of each subband in the sideband signals in the described code stream;
48kbps layer coding module, be used for by adopting the modulated lapped transform (mlt) scrambler that the stereosonic sideband signals in broadband is carried out the embedded encoded of 16kbps code check, obtain the code stream between 32kbps~48kbps code check, comprise the modulated lapped transform (mlt) coefficient quantization index information of time important subband in the sideband signals in the described code stream.
14, stereo encoding apparatus according to claim 10 is characterized in that, when selecting ultra broadband stereo coding pattern,
Described med signal coding unit, being further used for the stereosonic med signal of ultra broadband is carried out the all-key rate is the embedded encoded of 48kbps, forms the code stream between 0~48kbps code check;
Described sideband signals coding unit is further used for the stereosonic sideband signals of ultra broadband is carried out the embedded encoded of 16kbps code check, forms the code stream between 48kbps~64kbps code check.
15, stereo encoding apparatus according to claim 14 is characterized in that, described med signal coding unit comprises:
Med signal low frequency coding module, being used for the low frequency part of the stereosonic med signal of ultra broadband is carried out the all-key rate is the embedded encoded of 32kbps, obtains the code stream between 0~32kbps code check;
Med signal high-frequency coding module is used for the HFS of the stereosonic med signal of ultra broadband is carried out the embedded encoded of 16kbps code check, obtains the code stream between 32kbps~48kbps code check.
16, stereo encoding apparatus according to claim 14 is characterized in that, described sideband signals coding unit comprises:
Sideband signals low frequency coding module is used for the low frequency part of the stereosonic sideband signals of ultra broadband is carried out the embedded encoded of 8kbps code check, obtains the code stream between 48kbps~56kbps code check;
Sideband signals high-frequency coding module is used for the HFS of the stereosonic sideband signals of ultra broadband is carried out the embedded encoded of 8kbps code check, obtains the code stream between 56kbps~64kbps code check.
17, stereo encoding apparatus according to claim 15 is characterized in that, when being the ultra broadband stereophonic signal of 32kHz sampling rate, described med signal low frequency coding module comprises when stereo:
G.729.1 scrambler, be used for by adopt the embedded type speech encoding standard G.729.1 scrambler obtain the code stream between 0~32kbps code check to 2 sampling 16kHz and carry out the embedded encoded of 32kbps under the med signal of the stereosonic 0~7kHz frequency band of ultra broadband.
18, stereo encoding apparatus according to claim 15 is characterized in that, when being the ultra broadband stereophonic signal of 32kHz sampling rate, described med signal high-frequency coding module comprises when stereo:
40kbps layer coding module, be used for by adopting modulated lapped transform (mlt) that the med signal of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out the embedded encoded of 16kbps code check, obtain the code stream between 32kbps~48kbps code check, comprise the amplitude envelops quantitative information of interior each subband of 7kHz~14kHz frequency band range of med signal, the modulated lapped transform (mlt) coefficient quantization index information of classify control information and important subband in the described code stream;
48kbps layer coding module, be used for by adopting modulated lapped transform (mlt) that the med signal of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out the embedded encoded of 16kbps code check, obtain the code stream between 32kbps~48kbps code check, comprise the modulated lapped transform (mlt) coefficient quantization index information of the interior time important subband of 7kHz~14kHz frequency band range of med signal in the described code stream.
19, stereo encoding apparatus according to claim 16 is characterized in that, when being the ultra broadband stereophonic signal of 32kHz sampling rate, described sideband signals low frequency coding module comprises when stereo:
56kbps layer coding module, be used for by adopting modulated lapped transform (mlt) that the sideband signals of the stereosonic 0~7kHz frequency band of ultra broadband is carried out the embedded encoded of 8kbps code check, obtain the code stream between 48kbps~56kbps code check, comprise the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and all subbands of each subband in sideband signals 0~7kHz frequency band range in the described code stream.
20, stereo encoding apparatus according to claim 16 is characterized in that, when being the ultra broadband stereophonic signal of 32kHz sampling rate, described sideband signals high-frequency coding module comprises when stereo:
64kbps layer coding module, be used for by adopting modulated lapped transform (mlt) that the sideband signals of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out the embedded encoded of 8kbps code check, obtain the code stream between 56kbps~64kbps code check, comprise the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and all subbands of each subband in sideband signals 7kHz~14kHz frequency band range in the described code stream.
21, a kind of stereo decoding method is characterized in that, may further comprise the steps:
Sampling rate to the code stream of damascene structures is judged, selects corresponding decoding schema;
According to the decoding schema of selecting, the code rate layer that the code stream selection of damascene structures is different is decoded, solve med signal and sideband signals respectively;
Med signal that decoding is obtained and sideband signals are sued for peace and are asked poor, try to achieve stereosonic left and right acoustic channels, the synthetic stereophonic signal that obtains.
22, stereo decoding method according to claim 21, it is characterized in that, when selecting broadband stereo decoding pattern, described the different code rate layer of the code stream selection of damascene structures is decoded, solve med signal respectively and the sideband signals step comprises:
To the code stream between 0~32kbps code check, carrying out the all-key rate is the embedded decoding of 32kbps, obtains the stereosonic med signal in broadband;
To the code stream between 32kbps~48kbps code check, carry out the embedded decoding of 16kbps code check, obtain the stereosonic sideband signals in broadband.
23, stereo decoding method according to claim 22, it is characterized in that, code stream for the broadband stereophonic signal of 16kHz sampling rate, to the code stream between 0~32kbps code check, by adopt the embedded type speech encoding standard G.729.1 demoder carry out the embedded decoding that the all-key rate is 32kbps, obtain the stereosonic med signal in broadband.
24, stereo decoding method according to claim 22, it is characterized in that, code stream for the broadband stereophonic signal of 16kHz sampling rate, to the code stream between 32kbps~48kbps code check, by adopting the overlapping inverse transformation demoder of modulation to carry out the embedded decoding of 16kbps code check, obtain the stereosonic sideband signals in broadband; The overlapping inverse transformation demoder of described modulation is decoded to the code stream of different length, and sorting from high to low according to sub belt energy preferentially decodes the information of important subband in the code stream.
25, stereo decoding method according to claim 24, it is characterized in that, if decoding end is received the 40kbps code check, then decoding obtains the modulated lapped transform (mlt) coefficient of the important subband in the stereosonic sideband signals 0~7kHz frequency band range of broadband, to the modulated lapped transform (mlt) coefficient of inferior important subband in 0~7kHz frequency band range by adopting the generation of noise filling technique.
26, stereo decoding method according to claim 22 is characterized in that,
When code check was 48kbps, complete decoding went out stereosonic med signal in broadband and sideband signals;
When code check is 40kbps, decode stereosonic med signal in broadband and the stereosonic sideband signals in broadband that contains important sub-band information;
When code check smaller or equal to 32kbps, can decode the stereosonic med signal in broadband.
27, stereo decoding method according to claim 21, it is characterized in that, when selecting ultra broadband stereo decoding pattern, describedly to decode for the different code rate layer of the code stream selection of damascene structures, the step that solves med signal and sideband signals respectively comprises:
To the code stream between 0~32kbps code check, carrying out the all-key rate is the embedded decoding of 32kbps, obtains the med signal of the stereosonic 0~7kHz frequency band of ultra broadband;
To the code stream between 32kbps~48kbps code check, carry out the embedded decoding of 16kbps code check, obtain the med signal of the stereosonic 7kHz of ultra broadband~14kHz frequency band;
To the code stream between 48kbps~56kbps code check, carry out the embedded decoding of 8kbps code check, obtain the sideband signals of the stereosonic 0~7kHz frequency band of ultra broadband;
To the code stream between 56kbps~64kbps code check, carry out the embedded decoding of 8kbps code check, obtain the sideband signals of the stereosonic 7kHz of ultra broadband~14kHz frequency band.
28, stereo decoding method according to claim 27, it is characterized in that, code stream for the ultra broadband stereophonic signal of 32kHz sampling rate, to the code stream between 0~32kbps code check, by adopt the embedded type speech encoding standard G.729.1 demoder carry out the embedded decoding that the all-key rate is 32kbps, obtain the med signal of the stereosonic 0~7kHz frequency band of ultra broadband.
29, stereo decoding method according to claim 27, it is characterized in that, code stream for the ultra broadband stereophonic signal of 32kHz sampling rate, to the code stream between 32kbps~48kbps code check, by adopting the overlapping inverse transformation demoder of modulation to carry out the embedded decoding of 16kbps code check, obtain the med signal of the stereosonic 0~7kHz frequency band of ultra broadband; To the code stream between 48kbps~56kbps code check, carry out the embedded decoding of 8kbps code check by adopting the overlapping inverse transformation demoder of modulation, obtain the sideband signals of the stereosonic 0~7kHz frequency band of ultra broadband; To the code stream between 56kbps~64kbps code check, carry out embedded decoding by adopting the modulated lapped transform (mlt) demoder, obtain the sideband signals of the stereosonic 7kHz of ultra broadband~14kHz frequency band; The overlapping inverse transformation demoder of described modulation is decoded to the code stream of different length, and sorting from high to low according to sub belt energy preferentially decodes the information of important subband in the code stream.
30, stereo decoding method according to claim 29, it is characterized in that, if decoding end is received the 40kbps code check, then decoding obtains the modulated lapped transform (mlt) coefficient of the important subband in the stereosonic med signal 7kHz of ultra broadband~14kHz frequency band range, and the modulated lapped transform (mlt) coefficient of inferior important subband in 7kHz~14kHz frequency band range is generated roughly by employing noise filling technique.
31, stereo decoding method according to claim 27 is characterized in that,
When code check was 64kbps, complete decoding went out stereosonic med signal of ultra broadband and the stereosonic sideband signals of ultra broadband that comprises 0~14kHz band information;
When code check is 56kbps, decode the sideband signals of stereosonic med signal of ultra broadband and the stereosonic 0~7kHz frequency band of ultra broadband;
When code check is 48kbps, decode the stereosonic med signal of ultra broadband;
When code check during, decode the med signal of the stereosonic 0~7kHz frequency band of ultra broadband smaller or equal to 32kbps.
32, a kind of stereo decoding apparatus is characterized in that, comprising:
The decoding schema selected cell is used for the sampling rate of the code stream of damascene structures is judged, selects corresponding decoding schema;
The code rate layer selected cell is used for according to the decoding schema of selecting, the code rate layer different to the code stream selection of damascene structures;
The med signal decoding unit is used for the code rate layer decoding of selecting is obtained stereosonic med signal;
The sideband signals decoding unit is used for the code rate layer decoding of selecting is obtained stereosonic sideband signals;
The stereophonic signal synthesis unit is used for med signal that decoding is obtained and sideband signals and sues for peace and ask poor, tries to achieve stereosonic left and right acoustic channels, the synthetic stereophonic signal that obtains.
33, stereo decoding apparatus according to claim 32 is characterized in that, when selecting broadband stereo decoding pattern,
Described med signal decoding unit is further used for the code stream between 0~32kbps code check is carried out the embedded decoding that the all-key rate is 32kbps, obtains the stereosonic med signal in broadband;
Described sideband signals decoding unit is further used for the code stream between 32kbps~48kbps code check is carried out the embedded decoding of 16kbps code check, obtains the stereosonic sideband signals in broadband.
34, stereo decoding apparatus according to claim 33 is characterized in that, for the code stream of the broadband stereophonic signal of 16kHz sampling rate, described med signal decoding unit comprises:
G.729.1 demoder is used for the code stream between 0~32kbps code check, by adopt the embedded type speech encoding standard G.729.1 demoder carry out the embedded decoding that the all-key rate is 32kbps, obtain the stereosonic med signal in broadband.
35, stereo decoding apparatus according to claim 33 is characterized in that, for the code stream of the broadband stereophonic signal of 16kHz sampling rate, described sideband signals decoding unit comprises:
40kbps layer decoder module, be used for when code check is 40kbps, the embedded decoding of carrying out the 16kbps code check according to described 40kbps layer bit stream information obtains comprising the modulated lapped transform (mlt) coefficient of important subband, and, the modulated lapped transform (mlt) coefficient after filling is modulated the sideband signals that overlapping inverse transformation obtains the broadband stereophonic signal by adopting the noise filling technique that the modulated lapped transform (mlt) coefficient of the 48kbps decoding layer that lacks in the code stream is filled;
48kbps layer decoder module, be used for when code check is 48kbps, carry out the modulated lapped transform (mlt) coefficient of whole subbands that the embedded decoding of 16kbps code check obtains according to described 40kbps and the two-layer code stream information of 48kbps, modulate the sideband signals that overlapping inverse transformation obtains the broadband stereophonic signal.
36, stereo decoding apparatus according to claim 32 is characterized in that, when selecting ultra broadband stereo decoding pattern,
Described med signal decoding unit, being further used for the code stream between 0~48kbps code check is carried out the all-key rate is the embedded decoding of 48kbps, obtains the stereosonic med signal of ultra broadband;
Described sideband signals decoding unit is further used for the code stream between 48kbps~64kbps code check is carried out the embedded decoding of 16kbps code check, obtains the stereosonic sideband signals of ultra broadband.
37, stereo decoding apparatus according to claim 36 is characterized in that, described med signal decoding unit comprises:
Med signal low frequency decoder module is used for the code stream between 0~32kbps code check is carried out the embedded decoding that the all-key rate is 32kbps, obtains the med signal of the stereosonic 0~7kHz frequency band of ultra broadband;
Med signal high frequency decoder module is used for the code stream between 32kbps~48kbps code check is carried out the embedded decoding of 16kbps code check, obtains the med signal of the stereosonic 7kHz of ultra broadband~14kHz frequency band.
38, stereo decoding apparatus according to claim 36 is characterized in that, described sideband signals decoding unit comprises:
Sideband signals low frequency decoder module is used for the code stream between 48kbps~56kbps code check is carried out the embedded decoding of 8kbps code check, obtains the sideband signals of the stereosonic 0~7kHz frequency band of ultra broadband;
Sideband signals high frequency decoder module is used for the code stream between 56kbps~64kbps code check is carried out the embedded decoding of 8kbps code check, obtains the sideband signals of the stereosonic 7kHz of ultra broadband~14kHz frequency band.
According to the described stereo decoding apparatus of claim 37, it is characterized in that 39, for the code stream of the ultra broadband stereophonic signal of 32kHz sampling rate, described med signal low frequency decoder module comprises:
G.729.1 demoder, be used for the code stream between 0~32kbps code check, by adopt the embedded type speech encoding standard G.729.1 demoder carry out the embedded decoding that the all-key rate is 32kbps, last 2 sample 32kHz, obtain the med signal of the stereosonic 0~7kHz frequency band of ultra broadband
According to the described stereo decoding apparatus of claim 37, it is characterized in that 40, for the code stream of the ultra broadband stereophonic signal of 32kHz sampling rate, described med signal high frequency decoder module comprises:
40kbps layer decoder module, be used for when code check is 40kbps, the embedded decoding of carrying out the 16kbps code check according to described 40kbps layer bit stream information obtains comprising the modulated lapped transform (mlt) coefficient of important subband, and, the modulated lapped transform (mlt) coefficient after filling is modulated the med signal that overlapping inverse transformation obtains the stereosonic 7kHz~14kHz of ultra broadband by adopting the noise filling technique that the modulated lapped transform (mlt) coefficient of the 48kbps decoding layer that lacks in the code stream is filled;
48kbps layer decoder module, be used for when code check is 48kbps, the embedded decoding of carrying out the 16kbps code check according to described 40kbps and the two-layer code stream information of 48kbps obtains comprising the modulated lapped transform (mlt) coefficient of whole subbands, modulates the med signal that overlapping inverse transformation obtains the stereosonic 7kHz of ultra broadband~14kHz frequency band.
According to the described stereo decoding apparatus of claim 38, it is characterized in that 41, for the code stream of the ultra broadband stereophonic signal of 32kHz sampling rate, described sideband signals low frequency decoder module comprises:
56kbps layer decoder module, be used for when code check is 56kbps, the embedded decoding of carrying out the 8kbps code check according to 56kbps layer bit stream information obtains comprising the modulated lapped transform (mlt) coefficient of whole subbands, modulates the sideband signals that overlapping inverse transformation obtains the stereosonic 0~7kHz frequency band of ultra broadband
Figure A2008100944360011C2
According to the described stereo decoding apparatus of claim 38, it is characterized in that 42, for the code stream of the ultra broadband stereophonic signal of 32kHz sampling rate, described sideband signals high frequency decoder module comprises:
64kbps layer decoder module, be used for when code check is 64kbps, the embedded decoding of carrying out the 8kbps code check according to 64kbps layer bit stream information obtains comprising the modulated lapped transform (mlt) coefficient of whole subbands, modulates the sideband signals that overlapping inverse transformation obtains the stereosonic 7kHz of ultra broadband~14kHz frequency band.
43, a kind of stereo coding system is characterized in that, comprising:
Coding mode selected cell, the sampling rate that is used for stereophonic signal are judged, select the respective coding pattern;
The broadband stereo encoding apparatus is used for difference is averaged, asked to the stereosonic left and right acoustic channels in broadband, obtains the med signal and the sideband signals of broadband stereophonic signal respectively; Described med signal and sideband signals are carried out embedded encoded processing respectively, form the code stream of broadband damascene structures;
The ultra broadband stereo encoding apparatus is used for difference is averaged, asked to the stereosonic left and right acoustic channels of ultra broadband, obtains the med signal and the sideband signals of ultra broadband stereophonic signal respectively; Described med signal and sideband signals are carried out embedded encoded processing respectively, form the code stream of ultra broadband damascene structures.
According to the described stereo coding of claim 43 system, it is characterized in that 44, when selecting broadband stereo coding pattern, it is embedded encoded to select the broadband stereo encoding apparatus to carry out, described broadband stereo encoding apparatus comprises:
The med signal acquiring unit is used for the left and right acoustic channels of broadband stereophonic signal averaged and obtains the stereosonic med signal in broadband;
The sideband signals acquiring unit is used for asking difference to obtain the stereosonic sideband signals in broadband to the left and right acoustic channels of broadband stereophonic signal;
The med signal coding unit, being used for the stereosonic med signal in described broadband is carried out the all-key rate is the embedded encoded of 32kbps, obtains the code stream between 0~32kbps code check;
The sideband signals coding unit is used for the stereosonic sideband signals in described broadband is carried out the embedded encoded of 16kbps code check, obtains the code stream between 32kbps~48kbps code check.
According to the described stereo coding of claim 44 system, it is characterized in that 45, when the signal of input was the broadband stereophonic signal of 16kHz sampling rate, described med signal coding unit comprised:
G.729.1 scrambler, be used for by adopt the embedded type speech encoding standard G.729.1 scrambler the stereosonic med signal in described broadband is carried out 32kbps code check coding, obtain the code stream between 0~32kbps code check;
And described sideband signals coding unit comprises:
40kbps layer coding module, be used for by adopting the modulated lapped transform (mlt) scrambler that the stereosonic sideband signals in broadband is carried out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, and comprised the modulated lapped transform (mlt) coefficient quantization index information of the amplitude envelops quantitative information of each subband, classification control information and important subband in the sideband signals in the code stream;
48kbps layer coding module, be used for by adopting the modulated lapped transform (mlt) scrambler that the stereosonic sideband signals in broadband is carried out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, and included only the modulated lapped transform (mlt) coefficient quantization index information of time important subband in the sideband signals in the code stream.
According to the described stereo coding of claim 43 system, it is characterized in that 46, when selecting ultra broadband stereo coding pattern, it is embedded encoded to select the ultra broadband stereo encoding apparatus to carry out, described ultra broadband stereo encoding apparatus comprises:
The med signal acquiring unit is used for the left and right acoustic channels of ultra broadband stereophonic signal averaged and obtains the stereosonic med signal of ultra broadband;
The sideband signals acquiring unit is used for asking difference to obtain the stereosonic sideband signals of ultra broadband to the left and right acoustic channels of ultra broadband stereophonic signal;
The med signal coding unit, being used for the stereosonic med signal of described ultra broadband is carried out the all-key rate is the embedded encoded of 48kbps, obtains the code stream between 0~48kbps code check;
The sideband signals coding unit is used for the stereosonic sideband signals of ultra broadband is carried out the embedded encoded of 16kbps code check, obtains the code stream between 48kbps~64kbps code check.
47, according to the described stereo coding of claim 46 system, it is characterized in that, when the signal of input is the ultra broadband stereophonic signal of 32kHz sampling rate,
Described med signal coding unit comprises:
G.729.1 scrambler, be used for by adopt the embedded type speech encoding standard G.729.1 scrambler obtain the code stream between 0~32kbps code check to 2 sampling 16kHz to carry out the all-key rate be the embedded encoded of 32kbps under the med signal of the stereosonic 0~7kHz frequency band of ultra broadband;
40kbps layer coding module, be used for by adopting the modulated lapped transform (mlt) scrambler that the med signal of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, comprised the amplitude envelops quantitative information of interior each subband of 7kHz~14kHz frequency band range of med signal, the modulated lapped transform (mlt) coefficient quantization index information of classify control information and important subband in the described code stream;
48kbps layer coding module, be used for by adopting the modulated lapped transform (mlt) scrambler that the med signal of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out 16kbps code check coding, obtain the code stream between 32kbps~48kbps code check, included only the modulated lapped transform (mlt) coefficient quantization index information of the interior time important subband of 7kHz~14kHz frequency band range of med signal in the described code stream;
Described sideband signals coding unit comprises:
56kbps layer coding module, be used for by adopting the modulated lapped transform (mlt) scrambler that the sideband signals of the stereosonic 0~7kHz frequency band of ultra broadband is carried out 8kbps code check coding, obtain the code stream between 48kbps~56kbps code check, comprised the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and all subbands of each subband in sideband signals 0~7kHz frequency band range in the described code stream;
64kbps layer coding module, be used for by adopting the modulated lapped transform (mlt) scrambler that the sideband signals of the stereosonic 7kHz of ultra broadband~14kHz frequency band is carried out 8kbps code check coding, obtain the code stream between 56kbps~64kbps code check, comprised the modulated lapped transform (mlt) coefficient quantization index information of amplitude envelops quantitative information, classification control information and all subbands of each subband in sideband signals 7kHz~14kHz frequency band range in the described code stream.
48, a kind of stereo decoding system is characterized in that, comprising:
The decoding schema selected cell is used for the sampling rate of the code stream of damascene structures is judged, selects corresponding decoding schema;
The broadband stereo decoding apparatus, be used for the different code rate layer of the code stream selection of the stereosonic damascene structures in broadband is carried out embedded decoding, obtain stereosonic med signal in broadband and sideband signals respectively, suing for peace and asking to differ from obtains stereosonic left and right acoustic channels, the synthetic broadband stereophonic signal that obtains;
The ultra broadband stereo encoding apparatus, be used for the different code rate layer of the code stream selection of the stereosonic damascene structures of ultra broadband is carried out embedded decoding, obtain stereosonic med signal of ultra broadband and sideband signals respectively, suing for peace and asking to differ from obtains stereosonic left and right acoustic channels, the synthetic ultra broadband stereophonic signal that obtains.
According to the described stereo decoding of claim 48 system, it is characterized in that 49, when selecting broadband stereo decoding pattern, select the broadband stereo decoding apparatus to carry out embedded decoding, described broadband stereo decoding apparatus comprises:
The code rate layer selected cell is used for the code rate layer different to the code stream selection of the stereosonic damascene structures in broadband;
The med signal decoding unit is used for the code stream between 0~32kbps code check is carried out the embedded decoding that the all-key rate is 32kbps, obtains the stereosonic med signal in broadband;
The sideband signals decoding unit is used for the code stream between 32kbps~48kbps code check is carried out the embedded decoding of 16kbps code check, obtains the stereosonic sideband signals in broadband;
The stereophonic signal synthesis unit is used for stereosonic med signal in broadband that decoding is obtained and sideband signals and sues for peace and ask poor, tries to achieve stereosonic left and right acoustic channels respectively, the synthetic broadband stereophonic signal that obtains.
According to the described stereo decoding of claim 49 system, it is characterized in that 50, for the code stream of the broadband stereophonic signal of 16kHz sampling rate, described med signal decoding unit comprises:
G.729.1 demoder is used for the code stream between 0~32kbps code check is carried out the embedded decoding that the all-key rate is 32kbps, obtains the stereosonic med signal in broadband;
And described sideband signals decoding unit comprises:
40kbps layer decoder module, be used for when code check is 40kbps, the embedded decoding of carrying out the 16kbps code check according to described 40kbps layer bit stream information obtains comprising the modulated lapped transform (mlt) coefficient of important subband, and, the modulated lapped transform (mlt) coefficient after filling is modulated the sideband signals that overlapping inverse transformation obtains the broadband stereophonic signal by adopting the noise filling technique that the modulated lapped transform (mlt) coefficient of the 48kbps decoding layer that lacks in the code stream is filled;
48kbps layer decoder module, be used for when code check is 48kbps, the embedded decoding of carrying out the 16kbps code check according to described 40kbps and the two-layer code stream information of 48kbps obtains comprising the modulated lapped transform (mlt) coefficient of whole subbands, modulates the sideband signals that overlapping inverse transformation obtains the broadband stereophonic signal.
According to the described stereo decoding of claim 48 system, it is characterized in that 51, when selecting ultra broadband stereo decoding pattern, select the ultra broadband stereo decoding apparatus to carry out embedded decoding, described ultra broadband stereo decoding apparatus comprises:
The code rate layer selected cell is used for the code rate layer different to the code stream selection of the stereosonic damascene structures of ultra broadband;
The med signal decoding unit, being used for the code stream between 0~48kbps code check is carried out the all-key rate is the embedded decoding of 48kbps, obtains the stereosonic med signal of ultra broadband;
The sideband signals decoding unit is used for the code stream between 48kbps~64kbps code check is carried out the embedded decoding of 16kbps code check, obtains the stereosonic sideband signals of ultra broadband;
The stereophonic signal synthesis unit is used for stereosonic med signal of ultra broadband that decoding is obtained and sideband signals and sues for peace and ask poor, tries to achieve stereosonic left and right acoustic channels respectively, the synthetic ultra broadband stereophonic signal that obtains.
52, according to the described stereo decoding of claim 51 system, it is characterized in that, for the code stream of the ultra broadband stereophonic signal of 32kHz sampling rate,
Described med signal decoding unit comprises:
G.729.1 demoder, be used for the code stream between 0~32kbps code check, by adopt the embedded type speech encoding standard G.729.1 demoder carry out the embedded decoding that the all-key rate is 32kbps, last 2 sample 32kHz, obtain the med signal of the stereosonic 0~7kHz frequency band of ultra broadband;
40kbps layer decoder module, be used for when code check is 40kbps, the embedded decoding of carrying out the 16kbps code check according to described 40kbps layer bit stream information obtains comprising the modulated lapped transform (mlt) coefficient of important subband, and, the modulated lapped transform (mlt) coefficient after filling is modulated the med signal that overlapping inverse transformation obtains the stereosonic 7kHz~14kHz of ultra broadband by adopting the noise filling technique that the modulated lapped transform (mlt) coefficient of the 48kbps decoding layer that lacks in the code stream is filled;
48kbps layer decoder module, be used for when code check is 48kbps, the embedded decoding of carrying out 16kbp s code check according to described 40kbps and the two-layer code stream information of 48kbps obtains comprising the modulated lapped transform (mlt) coefficient of whole subbands, modulates the med signal that overlapping inverse transformation obtains the stereosonic 7kHz of ultra broadband~14kHz frequency band;
Described sideband signals decoding unit comprises:
56kbps layer decoder module, be used for when code check is 56kbps, the embedded decoding of carrying out the 8kbps code check according to 56kbps layer bit stream information obtains comprising the modulated lapped transform (mlt) coefficient of whole subbands, modulates the sideband signals that overlapping inverse transformation obtains the stereosonic 0~7kHz frequency band of ultra broadband;
64kbps layer decoder module, be used for when code check is 64kbps, the embedded decoding of carrying out the 8kbps code check according to 64kbps layer bit stream information obtains comprising the modulated lapped transform (mlt) coefficient of whole subbands, modulates the sideband signals that overlapping inverse transformation obtains the stereosonic 7kHz of ultra broadband~14kHz frequency band.
CNA2008100944366A 2008-04-30 2008-04-30 Stereo encoding and decoding method, a coder-decoder and encoding and decoding system Pending CN101572088A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNA2008100944366A CN101572088A (en) 2008-04-30 2008-04-30 Stereo encoding and decoding method, a coder-decoder and encoding and decoding system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNA2008100944366A CN101572088A (en) 2008-04-30 2008-04-30 Stereo encoding and decoding method, a coder-decoder and encoding and decoding system

Publications (1)

Publication Number Publication Date
CN101572088A true CN101572088A (en) 2009-11-04

Family

ID=41231421

Family Applications (1)

Application Number Title Priority Date Filing Date
CNA2008100944366A Pending CN101572088A (en) 2008-04-30 2008-04-30 Stereo encoding and decoding method, a coder-decoder and encoding and decoding system

Country Status (1)

Country Link
CN (1) CN101572088A (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101964202A (en) * 2010-09-09 2011-02-02 南京中兴特种软件有限责任公司 Audio data file playback processing method mixed with multiple encoded formats
CN102280107A (en) * 2010-06-10 2011-12-14 华为技术有限公司 Sideband residual signal generating method and device
CN103069721A (en) * 2010-08-24 2013-04-24 杜比国际公司 Reduction of spurious uncorrelation in fm radio noise
CN103854650A (en) * 2012-11-30 2014-06-11 中兴通讯股份有限公司 Stereo audio coding method and device
CN103971692A (en) * 2013-01-28 2014-08-06 北京三星通信技术研究有限公司 Audio processing method, device and system
CN106796798A (en) * 2014-07-28 2017-05-31 弗劳恩霍夫应用研究促进协会 Apparatus and method for filling generation enhancing signal using independent noise
CN107293303A (en) * 2017-06-16 2017-10-24 苏州蜗牛数字科技股份有限公司 A kind of multichannel voice lost packet compensation method
CN110462733A (en) * 2017-03-31 2019-11-15 华为技术有限公司 The decoding method and codec of multi-channel signal
CN112509591A (en) * 2020-12-04 2021-03-16 北京百瑞互联技术有限公司 Audio coding and decoding method and system

Cited By (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102280107A (en) * 2010-06-10 2011-12-14 华为技术有限公司 Sideband residual signal generating method and device
CN102280107B (en) * 2010-06-10 2013-01-23 华为技术有限公司 Sideband residual signal generating method and device
US9123329B2 (en) 2010-06-10 2015-09-01 Huawei Technologies Co., Ltd. Method and apparatus for generating sideband residual signal
CN103069721A (en) * 2010-08-24 2013-04-24 杜比国际公司 Reduction of spurious uncorrelation in fm radio noise
US9094754B2 (en) 2010-08-24 2015-07-28 Dolby International Ab Reduction of spurious uncorrelation in FM radio noise
CN103069721B (en) * 2010-08-24 2016-03-23 杜比国际公司 Reduce the pseudo-irrelevance in FM radio broadcast noise
CN101964202A (en) * 2010-09-09 2011-02-02 南京中兴特种软件有限责任公司 Audio data file playback processing method mixed with multiple encoded formats
CN101964202B (en) * 2010-09-09 2012-03-28 南京中兴特种软件有限责任公司 Audio data file playback processing method mixed with multiple encoded formats
CN103854650A (en) * 2012-11-30 2014-06-11 中兴通讯股份有限公司 Stereo audio coding method and device
CN103971692A (en) * 2013-01-28 2014-08-06 北京三星通信技术研究有限公司 Audio processing method, device and system
CN106796798A (en) * 2014-07-28 2017-05-31 弗劳恩霍夫应用研究促进协会 Apparatus and method for filling generation enhancing signal using independent noise
CN106796798B (en) * 2014-07-28 2021-03-05 弗劳恩霍夫应用研究促进协会 Apparatus and method for generating an enhanced signal using independent noise filling
US11264042B2 (en) 2014-07-28 2022-03-01 Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. Apparatus and method for generating an enhanced signal using independent noise-filling information which comprises energy information and is included in an input signal
US11705145B2 (en) 2014-07-28 2023-07-18 Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. Apparatus and method for generating an enhanced signal using independent noise-filling
US11908484B2 (en) 2014-07-28 2024-02-20 Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. Apparatus and method for generating an enhanced signal using independent noise-filling at random values and scaling thereupon
CN110462733A (en) * 2017-03-31 2019-11-15 华为技术有限公司 The decoding method and codec of multi-channel signal
US11386907B2 (en) 2017-03-31 2022-07-12 Huawei Technologies Co., Ltd. Multi-channel signal encoding method, multi-channel signal decoding method, encoder, and decoder
US11894001B2 (en) 2017-03-31 2024-02-06 Huawei Technologies Co., Ltd. Multi-channel signal encoding method, multi-channel signal decoding method, encoder, and decoder
CN107293303A (en) * 2017-06-16 2017-10-24 苏州蜗牛数字科技股份有限公司 A kind of multichannel voice lost packet compensation method
CN112509591A (en) * 2020-12-04 2021-03-16 北京百瑞互联技术有限公司 Audio coding and decoding method and system
CN112509591B (en) * 2020-12-04 2024-05-14 北京百瑞互联技术股份有限公司 Audio encoding and decoding method and system

Similar Documents

Publication Publication Date Title
CN101572088A (en) Stereo encoding and decoding method, a coder-decoder and encoding and decoding system
CN101253557B (en) Stereo encoding device and stereo encoding method
CN101842832B (en) Encoder and decoder
CN100559465C (en) The variable frame length coding that fidelity is optimized
CN101128866B (en) Optimized fidelity and reduced signaling in multi-channel audio encoding
CN1110145C (en) Scalable audio coding/decoding method and apparatus
JP4781153B2 (en) Audio data encoding and decoding apparatus, and audio data encoding and decoding method
CN101556799B (en) Audio decoding method and audio decoder
KR101221918B1 (en) A method and an apparatus for processing a signal
US20100322429A1 (en) Joint Enhancement of Multi-Channel Audio
CN103187065B (en) The disposal route of voice data, device and system
JPH11317672A (en) Stereophonic audio coding and decoding method/apparatus capable of bit-rate control
JPH11186911A (en) Audio encoding/decoding method capable of adjusting bit rate, device therefor and recording medium with the method recorded therein
WO2004023457A1 (en) Sound encoding apparatus and sound encoding method
CN1195160A (en) Scalable audio coding/decoding method and apparatus
JP2011509428A (en) Audio signal processing method and apparatus
CN102332267A (en) The scalable audio codec of full range band
CN113470667A (en) Voice signal coding and decoding method and device, electronic equipment and storage medium
US7835915B2 (en) Scalable stereo audio coding/decoding method and apparatus
CN101572087B (en) Method and device for encoding and decoding embedded voice or voice-frequency signal
US8024187B2 (en) Pulse allocating method in voice coding
US9830919B2 (en) Acoustic signal coding apparatus, acoustic signal decoding apparatus, terminal apparatus, base station apparatus, acoustic signal coding method, and acoustic signal decoding method
KR20060036724A (en) Method and apparatus for encoding/decoding audio signal
Chen et al. Scalefactor based bit shift FGS audio coding
Shen et al. A progressive algorithm for perceptual coding of digital audio signals

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C12 Rejection of a patent application after its publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20091104