CN102419978B - Audio decoder and frequency spectrum reconstructing method and device for audio decoding - Google Patents
Audio decoder and frequency spectrum reconstructing method and device for audio decoding Download PDFInfo
- Publication number
- CN102419978B CN102419978B CN2011102435396A CN201110243539A CN102419978B CN 102419978 B CN102419978 B CN 102419978B CN 2011102435396 A CN2011102435396 A CN 2011102435396A CN 201110243539 A CN201110243539 A CN 201110243539A CN 102419978 B CN102419978 B CN 102419978B
- Authority
- CN
- China
- Prior art keywords
- frequency spectrum
- data
- decoding
- code stream
- convergent
- 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.)
- Active
Links
- 238000001228 spectrum Methods 0.000 title claims abstract description 186
- 238000000034 method Methods 0.000 title claims abstract description 143
- 238000013139 quantization Methods 0.000 claims abstract description 125
- 238000012545 processing Methods 0.000 claims abstract description 93
- 230000008569 process Effects 0.000 claims abstract description 80
- 230000003595 spectral effect Effects 0.000 claims abstract description 60
- 101000591286 Homo sapiens Myocardin-related transcription factor A Proteins 0.000 claims description 6
- 102100034099 Myocardin-related transcription factor A Human genes 0.000 claims description 6
- 238000005457 optimization Methods 0.000 description 11
- 238000013500 data storage Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 7
- 238000011002 quantification Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 230000009467 reduction Effects 0.000 description 4
- 230000006835 compression Effects 0.000 description 3
- 238000007906 compression Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000005236 sound signal Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000010076 replication Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Images
Landscapes
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention relates to an audio decoder and a frequency spectrum reconstructing method and device for audio decoding. The frequency spectrum reconstructing method for audio decoding comprises the following steps of: carrying out entropy decoding on an input audio code stream, and directly carrying out inverse quantization and reverse scaling processing on obtained frequency spectrum data obtained after decoding; if the audio code stream is processed by adopting pulse data, carrying out pulse data reconstruction on the basis of output spectral lines subjected to the inverse quantization and reverse scaling processing, and carrying out inverse quantization and reverse scaling processing on frequency spectrum data obtained after the pulse data is reconstructed. According to the technical scheme of the invention, under the condition that the decoding processing accuracy is not affected, the operational complexity in the frequency spectrum reconstructing process of the audio decoding is reduced, and the decoding efficiency is improved.
Description
Technical field
The present invention relates to the audio encoding and decoding technique field, particularly the frequency spectrum reconfiguration method of a kind of audio decoder, audio decoder and device.
Background technology
Multimedia is comprehensive information resources, is the general designation of the media elements such as text, figure, sound, animation, video.At present, multimedia technology application gets more and more extensive.But how various media informations reduce these costs in storage, the cost processing and all will cost a lot of money during transmission, so as more convenient, process more quickly and transmit these data, be an emphasis in the multimedia application field.Therefore, the encoding and decoding technique of multi-medium data has become a gordian technique of the applications such as multimedia, network communication and computing machine.
Sound signal is a kind of of multi-medium data.The further investigation of human auditory system theory and the widespread use of digitizing technique, and market impels the digital audio decode technology to become an important research field of consumer electronics to the requirement of tending towards superiority of the happy quality of consumption assonance.At present, multiple audio coding (decoding) standard has appearred, for example: Advanced Audio Coding (AAC, Advanced Audio Coding), bit timeslice arithmetic coding (BSAC, Bit-Slice Arithmetic Coding), WMA (Windows Media Audio), MP3 (MPEG 1 audio layer 3), RA (Real Audio) etc.Especially AAC coding standard, it combines the advantage of multiple main flow audio decoding techniques, have signal compression than high, rebuild that tonequality is good, the characteristics such as encoding-decoding process high modularization and sound channel flexible configuration, become at present widely used a kind of audio coding standard, all be applied at numerous areas such as Digital Theater System, digital TV direct, digital audio broadcasting, teleconference system, radio communication and network multimedia business.The AAC coding techniques comes across 1997, is based at first the audio decoding techniques of MPEG-2.By Fraunhofer IIS, Dolby Laboratories, AT﹠amp; The company such as T, Sony joint development, purpose are to replace MP3 format.2000, the MPEG-4 standard is put into effect, other technology that AAC is again integrated, parameter stereo (PS for example, parametric stereo), frequency range copies (SBR, Spectral Band Replication) etc., for being different from traditional MPEG-2AAC, be called again MPEG-4AAC therefore will contain the AAC of SBR or PS characteristic.
Although the coding standards such as AAC have been widely used in hand-hold mobile device (such as mobile phone, multimedia player etc.), but because the resource-constrained (performance of processor, the amount of capacity of internal memory etc.) of hand-hold mobile device, decoding complex degree to coded datas such as AAC is relatively more responsive, wishes that namely decoding complex degree can be low as far as possible.In the implementation of decoding through the voice data (with the form of code stream) of the codings such as AAC, entropy decoding (Entropy Decoding), inverse quantization are processed, anti-convergent-divergent is processed and taken very large calculated amount, generally can take the closely complexity of half, therefore the optimization of above-mentioned implementation procedure be seemed particularly important.
Take the frequency spectrum reconfiguration of AAC audio decoder as example, common way generally is to do first entropy decoding (specifically being generally Hofmann decoding), (if necessary) then carry out pulse data reconstruct (pulse data re-construct), carry out at last that inverse quantization is processed and anti-convergent-divergent processes to realize frequency spectrum reconfiguration to audio code stream.Fig. 1 is the schematic flow sheet of frequency spectrum reconfiguration method of the audio decoder of prior art.As shown in Figure 1, the frequency spectrum reconfiguration method of prior art sound intermediate frequency decoding comprises: execution in step S101 at first, carry out the entropy decoding to the audio code stream of input; Then execution in step S102 judges whether described audio code stream adopts pulse data to process, otherwise execution in step S104, inverse quantization is processed, and step S105, and anti-convergent-divergent is processed, then to carry out step S103, pulse data reconstruct, then execution in step S104 and step S105.Need to prove, in the frequency spectrum reconfiguration flow process of above-mentioned audio decoder, after the step S101, usually the frequency spectrum data that parses after decoding through entropy can be stored among the buffer, when needs execution in step S103, can from described buffer, load the frequency spectrum data that parses and carry out pulse data reconstruct, to be stored in through the frequency spectrum data after the pulse data reconstruct described buffer more afterwards, during follow-up execution in step S104, the frequency spectrum data that obtains behind the load pulses data reconstruction from described buffer again carries out inverse quantization to be processed, the spectral line data that will obtain after will processing through inverse quantization is stored in described buffer, then during execution in step S105, from described buffer, load again spectral line data after inverse quantization is processed and carry out anti-convergent-divergent and process; When judging by step S102 when not needing execution in step S103, then step S104 loads from described buffer through the frequency spectrum data that parses after the entropy decoding and carries out the inverse quantization processing, and the step S105 spectral line data that then loading is stored after processing through inverse quantization from described buffer is again carried out anti-convergent-divergent and processed.In addition, step S101 carries out the just symbol of known at that time spectral line of entropy when decoding, does not need to judge, if yet need execution in step S103, then when the inverse quantization of execution in step S104 is processed, still need to rejudge the symbol of spectral line, therefore have the problem of double counting.Hence one can see that, in the frequency spectrum reconfiguration process of the audio decoder of prior art, can exist repeatedly the data storage to load and larger calculating redundancy, and computational complexity increases, decoding efficiency reduces thereby make.
Correlation technique can be the international patent application of US2007027677A1 with reference to publication number also, this Patent Application Publication a kind of implementation method of audio codec, its computation complexity is low, memory space can guarantee the high audio compression quality again simultaneously less, is applied to the Audio compression coding technology of handheld device, SOC (system on a chip) or special IC product and embedded system.
Summary of the invention
The problem that the present invention solves provides a kind of frequency spectrum reconfiguration method and device of audio decoder, and in the situation of the precision that does not affect the decoding processing, the computational complexity in the frequency spectrum reconfiguration process of reduction audio decoder improves decoding efficiency.
For addressing the above problem, technical scheme of the present invention provides a kind of frequency spectrum reconfiguration method of audio decoder, comprising:
Audio code stream to input carries out the entropy decoding, and the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent;
If described audio code stream employing pulse data is processed, then carry out pulse data reconstruct based on the spectral line data of output after processing through inverse quantization, anti-convergent-divergent, the frequency spectrum data that obtains after the described pulse data reconstruct is carried out inverse quantization, the processing of anti-convergent-divergent.
Optionally, described spectral line data based on output after processing through inverse quantization, anti-convergent-divergent is carried out pulse data reconstruct, to the frequency spectrum data that obtains after the described pulse data reconstruct carry out inverse quantization, anti-convergent-divergent is processed and is comprised:
After processing through inverse quantization, anti-convergent-divergent, take out the spectral line data that to carry out pulse data reconstruct the spectral line data of output;
Described spectral line data to taking-up quantizes, convergent-divergent is processed;
The frequency spectrum data that quantizes, convergent-divergent is exported after processing is carried out pulse data reconstruct;
To carry out inverse quantization, the processing of anti-convergent-divergent through the frequency spectrum data that obtains after the pulse data reconstruct.
Optionally, describedly process to be based on and determine that the needs that take out carry out that the symbol of the spectral line data of pulse data reconstruct realizes carry out inverse quantization, anti-convergent-divergent through the frequency spectrum data that obtains after the pulse data reconstruct.
Optionally, whether described audio code stream adopts pulse data to process to be the key word by obtaining in the described audio code stream to judge.
Optionally, the encoding and decoding standard of described audio code stream is AAC.
Optionally, described entropy is decoded as Hofmann decoding.
For addressing the above problem, technical scheme of the present invention also provides a kind of frequency spectrum reconfiguration method of audio decoder, comprise: if the audio code stream of input does not adopt pulse data to process, then described audio code stream is carried out the entropy decoding, the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent.
Optionally, described frequency spectrum reconfiguration method also comprises: if the audio code stream of input adopts pulse data to process, then described audio code stream is carried out the entropy decoding, the frequency spectrum data that parses is carried out pulse data reconstruct, to carry out inverse quantization, the processing of anti-convergent-divergent through the frequency spectrum data after the pulse data reconstruct.
Optionally, the encoding and decoding standard of described audio code stream is AAC, BSAC, WMA, MP3 or Real-Audio.
Optionally, described entropy is decoded as Hofmann decoding or arithmetic decoding.
For addressing the above problem, technical scheme of the present invention also provides a kind of frequency spectrum reconfiguration device of audio decoder, comprising:
The first frequency spectrum reconfiguration unit is suitable for the audio code stream of input is carried out the entropy decoding, and the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent;
Judging unit is suitable for judging whether described audio code stream adopts pulse data to process;
The second frequency spectrum reconfiguration unit, be suitable for when described audio code stream adopts pulse data to process, spectral line data based on output after processing through inverse quantization, anti-convergent-divergent is carried out pulse data reconstruct, and the frequency spectrum data that obtains after the described pulse data reconstruct is carried out inverse quantization, the processing of anti-convergent-divergent.
For addressing the above problem, technical scheme of the present invention also provides a kind of frequency spectrum reconfiguration device of audio decoder, comprising: judging unit is suitable for judging whether the audio code stream of input adopts pulse data to process; The 3rd frequency spectrum reconfiguration unit is suitable for when described judgment unit judges goes out described audio code stream and do not adopt pulse data to process described audio code stream being carried out the entropy decoding, with the frequency spectrum data that obtains after the decoding directly carry out inverse quantization, anti-convergent-divergent is processed.
Optionally, described frequency spectrum reconfiguration device also comprises the 4th frequency spectrum reconfiguration unit, be suitable for when described judgment unit judges goes out described audio code stream employing pulse data processing, described audio code stream is carried out the entropy decoding, the frequency spectrum data that obtains after the decoding is carried out pulse data reconstruct, to carry out inverse quantization, the processing of anti-convergent-divergent through the frequency spectrum data that obtains after the pulse data reconstruct.
For addressing the above problem, technical scheme of the present invention also provides a kind of audio decoder that comprises the frequency spectrum reconfiguration device of above-mentioned audio decoder.
Compared with prior art, the technical program has the following advantages:
Carry out the entropy decoding by the audio code stream to input, the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent, thus entropy decoding and inverse quantization, anti-convergent-divergent are processed and effectively integrated, avoided repeatedly data storage to load and larger calculating redundancy, thereby can be in the situation that do not affect the decoding processing accuracy, reduce computational complexity, improve decoding efficiency.
For the audio code stream that has adopted pulse data to process, the laggard horizontal pulse data reconstruction of spectral line data re-quantization, convergent-divergent of output after then will processing through inverse quantization, anti-convergent-divergent, the frequency spectrum data that obtains behind the paired pulses data reconstruction again carries out inverse quantization, anti-convergent-divergent is processed, even audio code stream has adopted the pulse data processing thus, also can process integrated basis realization to the frequency spectrum reconfiguration of audio code stream in entropy decoding and inverse quantization, anti-convergent-divergent, thereby reduce on the whole computational complexity, improve decoding efficiency.
Description of drawings
Fig. 1 is the schematic flow sheet of frequency spectrum reconfiguration method of the audio decoder of prior art;
Fig. 2 is the schematic flow sheet of the frequency spectrum reconfiguration method of the audio decoder that provides of the embodiment of the invention one;
Fig. 3 is the detailed process schematic diagram of step S203 among Fig. 2;
Fig. 4 is the structural representation of the frequency spectrum reconfiguration device of the audio decoder that provides of the embodiment of the invention one;
Fig. 5 is the schematic flow sheet of the frequency spectrum reconfiguration method of the audio decoder that provides of the embodiment of the invention two;
Fig. 6 is the schematic flow sheet of the frequency spectrum reconfiguration method of the audio decoder that provides of the embodiment of the invention three;
Fig. 7 is the structural representation of the frequency spectrum reconfiguration device of the audio decoder that provides of the embodiment of the invention three.
Embodiment
As stated in the Background Art, in the frequency spectrum reconfiguration process of the audio decoder of prior art, exist repeatedly the data storage to load and larger calculating redundancy, computational complexity increases, decoding efficiency reduces thereby make.The technical program is by processing and effectively integrate with the entropy decoding with inverse quantization, anti-convergent-divergent, avoided repeatedly data storage to load and larger calculating redundancy, thereby can be in the situation that do not affect the decoding processing accuracy, reduction computational complexity, raising decoding efficiency.
For above-mentioned purpose of the present invention, feature and advantage can more be become apparent, below in conjunction with accompanying drawing the specific embodiment of the present invention is described in detail.Set forth detail in the following description so that fully understand the present invention.But the present invention can be different from alternate manner described here and implements with multiple, and those skilled in the art can be in the situation that do similar popularization without prejudice to intension of the present invention.Therefore the present invention is not subjected to the restriction of following public embodiment.
Embodiment one
As stated in the Background Art, in the frequency spectrum reconfiguration process of the audio decoder of prior art, exist repeatedly data storages to load and larger calculating redundant, thereby cause that computational complexity increases, decoding efficiency is lower.Therefore, the inventor considers, can carry out integrated optimization by the step that entropy decoding, inverse quantization processing, anti-convergent-divergent are processed, and makes it carry out efficient higher.Integrated optimization described here comprises a plurality for the treatment of steps to data are merged on the one hand; Refer on the other hand after the processing of data through a treatment step, no longer be stored among the buffer for the result after processing (comprise processing after partial data or total data), but will directly process by next treatment step through the result after this step process (result after this processing temporarily is stored in the register of processor), after data are passed through the processing of integrated whole treatment steps, deposit the result that obtains at last in buffer.For a comparatively simple example explanation, suppose that original whole processing procedure comprises 2 treatment steps, be respectively a+b=c and d+c=e, carrying out so a+b result c afterwards can deposit in the buffer, when carrying out d+c, need from buffer, to load first c, then obtain e with the d addition again based on the c that loads; Through after integrated, whole processing procedure then becomes 1 treatment step, be a+b+d=e, whole processing procedure is directly finished by processor, certainly, the intermediate result of calculating is temporary in the register, does not repeat them here for the common practise of computer realm as for difference between register and the buffer.Need to prove, top only understands the integrated optimization of mentioning in the present embodiment for helping for example, should be as the restriction of the integrated optimization that the steps such as entropy decoding, inverse quantization processing, anti-convergent-divergent processing are carried out.
As the above analysis, one of advantage behind the integrated optimization is, has reduced the number of times that the data storage loads, thereby has improved the efficient that data are processed.Therefore, at first can consider the step that step and the anti-convergent-divergent of inverse quantization processing are processed is carried out integrated optimization, further, the step of the step that entropy can also be decoded and inverse quantization, anti-convergent-divergent is carried out integrated optimization.Owing to just can know the at that time symbol of spectral line when carrying out the entropy decoding, if the step of the step of entropy decoding and inverse quantization, anti-convergent-divergent is carried out integrated, be that inverse quantization, anti-convergent-divergent are directly carried out in the entropy decoding afterwards, the symbol of determined spectral line is directly finished the step of inverse quantization, anti-convergent-divergent in the time of then can utilizing the entropy decoding, and no longer need to rejudge the symbol of spectral line, thereby avoided the repetitive operation for the symbol decision of spectral line.Therefore, another advantage behind the integrated optimization is, reduced the calculating that the symbol decision because of spectral line causes redundant, improved operation efficiency.
Yet, for the audio code stream that adopts the AAC coding, can have the situation that adopts pulse data to process.In the AAC decoding standard, processing for large quantized value, the method that generally can adopt is to carry out pulse data to process, when using the method, large numerical value is deducted a difference and is become fractional value, then use the huffman table coding, the reduction of difference can be helped with a pulse structure in the back, namely carries out the operation of pulse data reconstruct.Therefore, for the frequency spectrum reconfiguration of audio code stream, except the step of entropy decoding, inverse quantization, anti-convergent-divergent, also comprise the step of pulse data reconstruct, this considers entropy decoding, pulse data reconstruct, inverse quantization and four steps of anti-convergent-divergent are carried out integrated optimization with regard to needs.Yet, can be with reference to figure 1, if want above-mentioned four steps are integrated, main trouble is to exist the step of pulse data reconstruction, because this step is to carry out computing to the data after quantizing, its special position has hindered the integrated of entropy decoding and inverse quantization, anti-convergent-divergent.
The frequency spectrum reconfiguration method of the audio decoder that the present embodiment provides has solved the technical matters that above-mentioned entropy decoding and inverse quantization, anti-convergent-divergent are difficult to be integrated optimization.Fig. 2 is the schematic flow sheet of the frequency spectrum reconfiguration method of the audio decoder that provides of the embodiment of the invention one.As shown in Figure 2, the frequency spectrum reconfiguration method of the audio decoder that provides of the present embodiment comprises:
Execution in step S201 carries out the entropy decoding to the audio code stream of inputting, and the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent;
Then execution in step S202 judges whether described audio code stream adopts pulse data to process;
Otherwise finished the frequency spectrum reconfiguration to described audio code stream, execution in step S203 then, spectral line data based on output after processing through inverse quantization, anti-convergent-divergent is carried out pulse data reconstruct, the frequency spectrum data that obtains after the described pulse data reconstruct is carried out inverse quantization, the processing of anti-convergent-divergent, thereby realize frequency spectrum reconfiguration.
In the present embodiment, the encoding and decoding standard of described audio code stream is AAC, for the audio code stream that adopts the AAC coding, can there be the situation that adopts pulse data to process, therefore need to judge whether described audio code stream adopts the step of pulse data processing, be execution in step S202, adopt pulse data to process if judge audio code stream, then also need to carry out pulse data reconstruct.In the present embodiment, described entropy decoding is specially Hofmann decoding, and in other embodiments, described entropy decoding also can be other entropy decoding method, for example arithmetic decoding.
The below elaborates to each step of said method.
At first, step S201 carries out the entropy decoding to the audio code stream of inputting, and the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent.Because described audio code stream is at output terminal (coding side) sound signal of sampling to be processed through quantification, convergent-divergent, and carry out just transferring to after the processing of entropy coding input end (decoding end), therefore, just need to carry out entropy decoding and inverse quantization to this audio code stream at input end, anti-convergent-divergent is processed the frequency spectrum reconfiguration that could realize sound signal.Need to prove, in order to reduce computational complexity, improve decoding efficiency, in the present embodiment, first no matter whether described audio code stream adopts pulse data to process, directly it is carried out the integrated computation of entropy decoding, inverse quantization, anti-convergent-divergent at step S201, then judge by step S202 whether described audio code stream adopts pulse data to process, otherwise having finished the frequency spectrum reconfiguration to described audio code stream, is then to carry out step S203 again.
" directly " described in the step S201 is in order to embody the integrated computation of entropy decoding, inverse quantization, anti-convergent-divergent, particularly, described audio code stream does not deposit first buffer in through the frequency spectrum data that entropy parses, but directly (this frequency spectrum data is stored in the register to the frequency spectrum data that parses, can directly carry out calculation process by microprocessor) carry out the processing of inverse quantization, anti-convergent-divergent, the spectral line data of exporting after the computing with inverse quantization, anti-convergent-divergent is stored in again carries out subsequent treatment in the buffer.
Among the step S202, judge that whether described audio code stream adopts pulse data to process is to realize by the key word that obtains in the described audio code stream.For example, whether to adopt the value of the key word that pulse data processes be 0 if get access to expression, represents that then audio code stream does not adopt pulse data to process; If the value of the described key word that gets access to is 1, represent that then audio code stream adopts pulse data to process, need to carry out to the frequency spectrum data that adopts pulse data to process the operation of pulse data reconstruct this moment.
Adopted the pulse data processing when judging described audio code stream by step S202, then carry out step S203, spectral line data based on output after processing through inverse quantization, anti-convergent-divergent is carried out pulse data reconstruct, the frequency spectrum data that obtains after the described pulse data reconstruct is carried out inverse quantization, the processing of anti-convergent-divergent, thereby realize frequency spectrum reconfiguration.Fig. 3 is the detailed process schematic diagram of step S203 among Fig. 2.Consult Fig. 3, step S203 specifically can comprise the steps:
S203a takes out the spectral line data that need to carry out pulse data reconstruct the spectral line data of output after processing through inverse quantization, anti-convergent-divergent;
S203b, to the described spectral line data of taking out quantize, convergent-divergent processes;
S203c carries out pulse data reconstruct to the frequency spectrum data that quantizes, convergent-divergent is exported after processing;
S203d is to carrying out inverse quantization, the processing of anti-convergent-divergent through the frequency spectrum data that obtains after the pulse data reconstruct.
Wherein, also comprise among the step S203a and determine that the needs that take out carry out the symbol of the spectral line data of pulse data reconstruct, specifically determine to carry out the symbol of the spectral line data of pulse data reconstruct by the mode that symbol is set.Because by entropy decoding and inverse quantization among the step S201, anti-convergent-divergent integrated computation, the symbol of determined spectral line is directly finished inverse quantization in the time of utilizing the entropy decoding, anti-convergent-divergent, pass through so inverse quantization, the symbol of the spectral line data of output was known after anti-convergent-divergent was processed, in subsequent step S203d, the frequency spectrum data that obtains after the reconstruct of process pulse data is carried out inverse quantization, anti-convergent-divergent repeats the computing of symbol decision when processing, then when taking out the spectral line data need to carry out pulse data reconstruct, step S203a just sets the symbol of corresponding spectral line data, in order to can realize the inverse quantization through the frequency spectrum data after the pulse data reconstruct based on set symbol when carrying out step S203d, anti-convergent-divergent is processed.
Particularly, for the setting of symbol, can do following computing:
If the needs that take out carry out the spectral line data spec of pulse data reconstruct＜=0, then: spec=-spec (get on the occasion of with carry out follow-up quantification, convergent-divergent is processed); Symbol sign=-1; Otherwise: sign=1.
Set after the symbol sign, spectral line data spec is made quantification, convergent-divergent processing, i.e. execution in step S203b.As previously mentioned, the step of pulse data reconstruct is just can carry out for the frequency spectrum data after quantizing, and therefore, need to re-start to the spectral line data spec that takes out quantification, convergent-divergent processing, specifically based on following formula:
Wherein, q is for re-starting the value of exporting after quantification, the convergent-divergent processing to spectral line data spec; Scf is coding subband zoom factor, is the given value of being imported into by (fixed length code is wherein resolved) in the entropy decoding step of front.
Obtain through quantification, convergent-divergent process after after the frequency spectrum data q of output, just can carry out pulse data reconstruct to q, i.e. execution in step S203c, specifically based on following formula:
q′＝q+pulse_amp
Wherein, q ' is the frequency spectrum data that obtains through after the pulse data reconstruct; Pulse_amp is current spectral line pulse height reconstruct offset, equally also is the given value of being imported into by entropy decoding step before (fixed length code is wherein resolved).
Computing through pulse data reconstruct obtains frequency spectrum data q ' afterwards, and then q ' is carried out inverse quantization, anti-convergent-divergent and process to realize frequency spectrum reconfiguration to audio code stream, i.e. execution in step S203d, specifically based on following formula:
Wherein, spec ' is the spectral line data of output after q ' is processed through inverse quantization, anti-convergent-divergent; Sign be before set symbol, the positive negative value that is used for the reduction spectral line, therefore can think, describedly process to be based on and determine that the needs that take out carry out that the symbol of the spectral line data of pulse data reconstruct realizes carry out inverse quantization, anti-convergent-divergent through the frequency spectrum data that obtains after the pulse data reconstruct.
So far, just finished frequency spectrum reconfiguration to described audio code stream.From the implementation of above-mentioned steps S203, in the present embodiment, increased much for the computation complexity of the step of pulse data reconstruct.Be not need to carry out S203b and these two steps of S203d in the implementation method of prior art, because when preparing to carry out the step of pulse data reconstruct, data stream also belongs to the data behind the quantization zooming, and the calculating of directly carrying out pulse data reconstruct gets final product.And in the present embodiment, when preparing to carry out the step of pulse data reconstruct, data stream has been the data after the anti-convergent-divergent of inverse quantization.Therefore, in order to do pulse data reconstruct, need to again do quantization zooming and process, namely the present embodiment is introduced the step of the heavy convergent-divergent of weightization.But because in the AAC coding standard, stipulate that each frame adopts the spectral line number of pulse data reconstruct to be no more than 4 points, the needed subband zoom factor of convergent-divergent all is known simultaneously, do not need to recomputate, therefore the complexity that the integrated computation for realizing the entropy decoding in the step S201 of complexity that pulse data reconstruct increases, inverse quantization, anti-convergent-divergent in the technical scheme of the present embodiment reduces, in general, remain the complexity that has greatly reduced computing, improved the efficient of decoding.
In a large amount of emulation experiments, be the monophony film source of 64k for the bit rate that does not adopt pulse data reconstruct, adopt the described frequency spectrum reconfiguration method of the present embodiment, compare (for example described technical scheme of background technology) with the scheme of prior art, reduce at least approximately 65% complexity; And be 64k for the bit rate that adopts pulse data reconstruct, the spectral line number of pulse data reconstruct is 4 monophony film source, adopts the described frequency spectrum reconfiguration method of the present embodiment, compares with the scheme of prior art, reduces at least approximately 60% complexity.Therefore, also confirm by emulation experiment: although increase to some extent for the computational complexity of realizing pulse data reconstruct in the present embodiment, still reduced generally calculated amount, improved decoding efficiency, and computational accuracy is not produced any impact.
Need to prove, in the calculating of the present embodiment step S201, Hofmann decoding algorithm, inverse quantization and the anti-convergent-divergent algorithm that uses do not limit, be applicable to the anti-Zoom method of all entropy encoding/decodings and inverse quantization, in step S203, S203b, the computing formula implementation method among S203c and the S203d does not limit yet.
Based on the frequency spectrum reconfiguration method of above-mentioned audio decoder, the present embodiment also provides a kind of frequency spectrum reconfiguration device of audio decoder.Fig. 4 is the structural representation of the frequency spectrum reconfiguration device of the audio decoder that provides of the embodiment of the invention one, as shown in Figure 4, the frequency spectrum reconfiguration device of the audio decoder that the present embodiment provides comprises: the first frequency spectrum reconfiguration unit 401, be suitable for the audio code stream of input is carried out the entropy decoding, the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent; Judging unit 402 is suitable for judging whether described audio code stream adopts pulse data to process; The second frequency spectrum reconfiguration unit 403, link to each other with described judging unit 402, be suitable for when described audio code stream adopts pulse data to process, spectral line data based on output after processing through inverse quantization, anti-convergent-divergent is carried out pulse data reconstruct, and the frequency spectrum data that obtains after the described pulse data reconstruct is carried out inverse quantization, the processing of anti-convergent-divergent.
During implementation, described the first frequency spectrum reconfiguration unit 401 can comprise: entropy decoding unit 401a is suitable for the audio code stream of input is carried out the entropy decoding; The anti-unit for scaling 401b of the first inverse quantization is connected with described entropy decoding unit 401a, is suitable for the frequency spectrum data that obtains after the described entropy decoding unit 401a decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent.Described the second frequency spectrum reconfiguration unit 403 comprises: acquiring unit 403a, take out the spectral line data that need to carry out pulse data reconstruct the spectral line data that is suitable for exporting after processing through the anti-unit for scaling 401b of described the first inverse quantization inverse quantization, anti-convergent-divergent; Quantization zooming unit 403b links to each other with described acquiring unit 403a, be suitable for to the described spectral line data that described acquiring unit 403a takes out quantize, convergent-divergent processes; Pulse data reconfiguration unit 403c, unit 403b links to each other with described quantization zooming, is suitable for the frequency spectrum data that described quantization zooming unit 403b quantizes, convergent-divergent is exported after processing is carried out pulse data reconstruct; The anti-unit for scaling 403d of the second inverse quantization links to each other with described pulse data reconfiguration unit 403c, is suitable for the frequency spectrum data that obtains after the pulse data reconstruct of carrying out through described pulse data reconfiguration unit 403c is carried out inverse quantization, the processing of anti-convergent-divergent.
Described the second frequency spectrum reconfiguration unit 403 also comprises symbol determining unit 403e, described symbol determining unit 403e links to each other with acquiring unit 403a, the anti-unit for scaling 403d of the second inverse quantization, is suitable for determining that needs that described acquiring unit 403a takes out carry out the symbol of the spectral line data of pulse data reconstruct; The anti-unit for scaling 403d of described the second inverse quantization processes to the inverse quantization that carries out through the frequency spectrum data that obtains after the pulse data reconstruct, anti-convergent-divergent that the symbol that is based on the determined spectral line data of described symbol determining unit 403e realizes.Described judging unit 402 judges that whether described audio code stream adopts pulse data to process is by obtaining the key word realization in the described audio code stream.
In addition, the present embodiment also provides a kind of audio decoder that comprises the frequency spectrum reconfiguration device of above-mentioned audio decoder.The frequency spectrum reconfiguration device of described audio decoder and the implementation of audio decoder can with reference to the frequency spectrum reconfiguration method of above-mentioned audio decoder, not repeat them here.
Embodiment two
What the present embodiment and embodiment one distinguished to some extent is, audio code stream to input described in the embodiment one carries out the entropy decoding, the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the step that anti-convergent-divergent is processed is to carry out before judging the step whether described audio code stream adopt pulse data to process, and in the present embodiment, described audio code stream to input carries out the entropy decoding, and the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the step that anti-convergent-divergent is processed is based on to judge carries out in the situation that described audio code stream adopts or do not adopt pulse data to process.
Fig. 5 is the schematic flow sheet of the frequency spectrum reconfiguration method of the audio decoder that provides of the embodiment of the invention two.As shown in Figure 5, the frequency spectrum reconfiguration method of the audio decoder that provides of the present embodiment comprises:
Step S301 judges whether the audio code stream of input adopts pulse data to process;
No matter the result after step S301 carries out how, the audio code stream that the present embodiment all will be carried out input carries out the entropy decoding, the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the step that anti-convergent-divergent is processed, judge the situation that adopts the pulse data processing and the situation that does not adopt pulse data to process in order to distinguish, do not adopt pulse data to process if judge, then this step is labeled as step S302a, adopt pulse data to process if judge, then this step is labeled as step S302b, step S302a is identical with the content that step S302b carries out, but, step S302a executes just and has realized frequency spectrum reconfiguration, and after the step S302b execution, also need execution in step S303, based on the process inverse quantization, the spectral line data of output was carried out pulse data reconstruct after anti-convergent-divergent was processed, and the frequency spectrum data that obtains after the described pulse data reconstruct is carried out inverse quantization, anti-convergent-divergent is processed, thereby realizes the frequency spectrum reconfiguration to audio code stream.
But the relevant enforcement of frequency spectrum reconfiguration method does not repeat them here described in the implementation reference example one of the frequency spectrum reconfiguration method of the audio decoder that the present embodiment provides.
Embodiment three
All be to judge first described audio code stream whether process by pulse data among the present embodiment and the embodiment two, then after judging described audio code stream and not adopting pulse data to process, audio code stream to input carries out entropy decoding, the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, anti-convergent-divergent process to realize frequency spectrum reconfiguration to audio code stream.Difference is to some extent, among the embodiment two after judging described audio code stream and adopting pulse data to process, still can carry out to the audio code stream of input the entropy decoding, the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent, the step of carrying out again afterwards pulse data reconstruct with the correlation technique among the embodiment one; And the present embodiment then carries out pulse data reconstruct according to conventional method in the prior art after judging the processing of described audio code stream employing pulse data.
Fig. 6 is the schematic flow sheet of the frequency spectrum reconfiguration method of the audio decoder that provides of the embodiment of the invention three.Consult Fig. 6, the frequency spectrum reconfiguration method of the audio decoder that the present embodiment provides comprises:
Step S601 judges whether the audio code stream of input adopts pulse data to process;
Otherwise execution in step S602 carries out the entropy decoding to described audio code stream, and the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent; Owing to not needing to carry out the operation of pulse data reconstruct, step S602 is finished and has namely realized frequency spectrum reconfiguration to described audio code stream.
Be execution in step S603 then, described audio code stream is carried out the entropy decoding; Then execution in step S604 carries out pulse data reconstruct with the frequency spectrum data that obtains after the decoding; After this execution in step S605 carries out inverse quantization, the processing of anti-convergent-divergent to the frequency spectrum data that obtains after the reconstruct of process pulse data.Can find out, step S603 to the implementation of step S605 be processing mode conventional in the prior art, but because when reality is implemented, the situation that employing pulse data coding is processed in most of film source is fewer (namely judging the situation that adopts pulse data to process less) also, therefore, in general, the complexity that relies on the integrated computation of entropy decoding among the step S602, inverse quantization, anti-convergent-divergent to reduce, still can greatly reduce the complexity of computing, improve the efficient of decoding.
In the present embodiment, the encoding and decoding standard of described audio code stream can be AAC, BSAC, WMA, MP3 or Real-Audio etc., and described entropy decoding is generally Hofmann decoding or arithmetic decoding.Need to prove, because at present except the AAC encoding and decoding standard, all do not relate to the situation that to carry out pulse data reconstruct in other encoding and decoding standards, therefore, during encoding and decoding standard beyond the encoding and decoding standard of described audio code stream is AAC, do not need to judge whether described audio code stream adopts pulse data to process, the integrated computation that only needs to carry out entropy decoding, inverse quantization, anti-convergent-divergent just can be realized the frequency spectrum reconfiguration to audio code stream.
But the relevant enforcement of the implementation reference example one of the frequency spectrum reconfiguration method of the audio decoder that the present embodiment provides and the frequency spectrum reconfiguration method described in the embodiment two does not repeat them here.
Based on the frequency spectrum reconfiguration method of above-mentioned audio decoder, the present embodiment also provides a kind of frequency spectrum reconfiguration device of audio decoder.Fig. 7 is the structural representation of the frequency spectrum reconfiguration device of the audio decoder that provides of the embodiment of the invention three, as shown in Figure 7, the frequency spectrum reconfiguration device of the audio decoder that the present embodiment provides comprises: judging unit 701 is suitable for judging whether the audio code stream of input adopts pulse data to process; The 3rd frequency spectrum reconfiguration unit 702, be connected with described judging unit 701, be suitable for when described judging unit 701 is judged described audio code stream and do not adopted pulse data to process, described audio code stream being carried out the entropy decoding, with the frequency spectrum data that obtains after the decoding directly carry out inverse quantization, anti-convergent-divergent is processed.
Described frequency spectrum reconfiguration device also comprises the 4th frequency spectrum reconfiguration unit 702, be connected with described judging unit 701, be suitable for when described judging unit 701 is judged described audio code stream employing pulse data processing, described audio code stream is carried out the entropy decoding, the frequency spectrum data that obtains after the decoding is carried out pulse data reconstruct, to carry out inverse quantization, the processing of anti-convergent-divergent through the frequency spectrum data that obtains after the pulse data reconstruct.
In the present embodiment, the encoding and decoding standard of described audio code stream is AAC, BSAC, WMA, MP3 or Real-Audio, and described entropy is decoded as Hofmann decoding or arithmetic decoding.
In addition, the present embodiment also provides a kind of audio decoder that comprises the frequency spectrum reconfiguration device of above-mentioned audio decoder.The frequency spectrum reconfiguration device of described audio decoder and the implementation of audio decoder can with reference to the frequency spectrum reconfiguration method of above-mentioned audio decoder, not repeat them here.
To sum up, the frequency spectrum reconfiguration method of the audio decoder that embodiment of the present invention provides and device have following beneficial effect at least:
Carry out the entropy decoding by the audio code stream to input, the frequency spectrum data that obtains after the decoding is directly carried out inverse quantization, the processing of anti-convergent-divergent, thus entropy decoding and inverse quantization, anti-convergent-divergent are processed and effectively integrated, avoided repeatedly data storage to load and larger calculating redundancy, thereby can be in the situation that do not affect the decoding processing accuracy, reduce computational complexity, improve decoding efficiency.
For the audio code stream that has adopted pulse data to process, the laggard horizontal pulse data reconstruction of spectral line data re-quantization, convergent-divergent of output after then will processing through inverse quantization, anti-convergent-divergent, the frequency spectrum data that obtains behind the paired pulses data reconstruction again carries out inverse quantization, anti-convergent-divergent is processed, even audio code stream has adopted the pulse data processing thus, also can process integrated basis realization to the frequency spectrum reconfiguration of audio code stream in entropy decoding and inverse quantization, anti-convergent-divergent, thereby reduce on the whole computational complexity, improve decoding efficiency.
Although the present invention with preferred embodiment openly as above; but it is not to limit the present invention; any those skilled in the art without departing from the spirit and scope of the present invention; can utilize method and the technology contents of above-mentioned announcement that technical solution of the present invention is made possible change and modification; therefore; every content that does not break away from technical solution of the present invention; to any simple modification, equivalent variations and modification that above embodiment does, all belong to the protection domain of technical solution of the present invention according to technical spirit of the present invention.
Claims (19)
1. the frequency spectrum reconfiguration method of an audio decoder is characterized in that, comprising:
Audio code stream to input carries out the entropy decoding, and the frequency spectrum data that obtains after the decoding is carried out inverse quantization, the processing of anti-convergent-divergent;
If described audio code stream adopts pulse data to process, then after processing through inverse quantization, anti-convergent-divergent, take out the spectral line data that to carry out pulse data reconstruct the spectral line data of output;
Described spectral line data to taking-up quantizes, convergent-divergent is processed;
The frequency spectrum data that quantizes, convergent-divergent is exported after processing is carried out pulse data reconstruct;
To carry out inverse quantization, the processing of anti-convergent-divergent through the frequency spectrum data that obtains after the pulse data reconstruct.
2. frequency spectrum reconfiguration method according to claim 1, it is characterized in that, describedly process to be based on and determine that the needs that take out carry out that the symbol of the spectral line data of pulse data reconstruct realizes carry out inverse quantization, anti-convergent-divergent through the frequency spectrum data that obtains after the pulse data reconstruct.
3. frequency spectrum reconfiguration method according to claim 1 is characterized in that, whether described audio code stream adopts pulse data to process is by obtaining the key word judgement in the described audio code stream.
4. frequency spectrum reconfiguration method according to claim 1 is characterized in that, the encoding and decoding standard of described audio code stream is AAC.
5. frequency spectrum reconfiguration method according to claim 1 is characterized in that, described entropy is decoded as Hofmann decoding.
6. the frequency spectrum reconfiguration method of an audio decoder is characterized in that, comprising:
If the audio code stream of input does not adopt pulse data to process, then described audio code stream is carried out the entropy decoding, the frequency spectrum data that obtains after the decoding is carried out inverse quantization, the processing of anti-convergent-divergent;
If the audio code stream of inputting adopts pulse data to process, then described audio code stream is carried out the entropy decoding, the frequency spectrum data that parses is carried out pulse data reconstruct, to carry out inverse quantization, the processing of anti-convergent-divergent through the frequency spectrum data after the pulse data reconstruct.
7. frequency spectrum reconfiguration method according to claim 6 is characterized in that, the encoding and decoding standard of described audio code stream is AAC, BSAC, WMA, MP3 or Real-Audio.
8. frequency spectrum reconfiguration method according to claim 6 is characterized in that, described entropy is decoded as Hofmann decoding or arithmetic decoding.
9. the frequency spectrum reconfiguration device of an audio decoder is characterized in that, comprising:
The first frequency spectrum reconfiguration unit is suitable for the audio code stream of input is carried out the entropy decoding, and the frequency spectrum data that obtains after the decoding is carried out inverse quantization, the processing of anti-convergent-divergent;
Judging unit is suitable for judging whether described audio code stream adopts pulse data to process;
The second frequency spectrum reconfiguration unit, be suitable for when described audio code stream adopts pulse data to process, spectral line data based on output after processing through inverse quantization, anti-convergent-divergent is carried out pulse data reconstruct, and the frequency spectrum data that obtains after the described pulse data reconstruct is carried out inverse quantization, the processing of anti-convergent-divergent.
10. frequency spectrum reconfiguration device according to claim 9 is characterized in that, described the first frequency spectrum reconfiguration unit comprises:
The entropy decoding unit is suitable for the audio code stream of input is carried out the entropy decoding;
The anti-unit for scaling of the first inverse quantization is suitable for the frequency spectrum data that obtains after the described entropy decoding unit decodes is carried out inverse quantization, the processing of anti-convergent-divergent.
11. frequency spectrum reconfiguration device according to claim 9 is characterized in that, described the second frequency spectrum reconfiguration unit comprises:
Acquiring unit takes out the spectral line data that need to carry out pulse data reconstruct the spectral line data that is suitable for exporting after processing through inverse quantization, anti-convergent-divergent;
The quantization zooming unit, be suitable for to the described spectral line data of taking out quantize, convergent-divergent processes;
The pulse data reconfiguration unit is suitable for the frequency spectrum data that quantizes, convergent-divergent is exported after processing is carried out pulse data reconstruct;
The anti-unit for scaling of the second inverse quantization is suitable for carry out inverse quantization, the processing of anti-convergent-divergent through the frequency spectrum data that obtains after the pulse data reconstruct.
12. frequency spectrum reconfiguration device according to claim 11 is characterized in that, described the second frequency spectrum reconfiguration unit also comprises the symbol determining unit, is suitable for determining that needs that described acquiring unit takes out carry out the symbol of the spectral line data of pulse data reconstruct; The anti-unit for scaling of described the second inverse quantization processes to the inverse quantization that carries out through the frequency spectrum data that obtains after the pulse data reconstruct, anti-convergent-divergent that the symbol that is based on the determined described spectral line data of described symbol determining unit realizes.
13. frequency spectrum reconfiguration device according to claim 9 is characterized in that, whether the described audio code stream of described judgment unit judges adopts pulse data to process is by obtaining the key word realization in the described audio code stream.
14. frequency spectrum reconfiguration device according to claim 9 is characterized in that, the encoding and decoding standard of described audio code stream is AAC.
15. frequency spectrum reconfiguration device according to claim 9 is characterized in that described entropy is decoded as Hofmann decoding.
16. the frequency spectrum reconfiguration device of an audio decoder is characterized in that, comprising:
Judging unit is suitable for judging whether the audio code stream of input adopts pulse data to process;
The 3rd frequency spectrum reconfiguration unit is suitable for when described judgment unit judges goes out described audio code stream and do not adopt pulse data to process described audio code stream being carried out the entropy decoding, with the frequency spectrum data that obtains after the decoding carry out inverse quantization, anti-convergent-divergent is processed;
The 4th frequency spectrum reconfiguration unit, be suitable for when described judgment unit judges goes out described audio code stream employing pulse data processing, described audio code stream is carried out the entropy decoding, the frequency spectrum data that obtains after the decoding is carried out pulse data reconstruct, to carry out inverse quantization, the processing of anti-convergent-divergent through the frequency spectrum data that obtains after the pulse data reconstruct.
17. frequency spectrum reconfiguration device according to claim 16 is characterized in that, the encoding and decoding standard of described audio code stream is AAC, BSAC, WMA, MP3 or Real-Audio.
18. frequency spectrum reconfiguration device according to claim 16 is characterized in that described entropy is decoded as Hofmann decoding or arithmetic decoding.
19. audio decoder that comprises the frequency spectrum reconfiguration device of each described audio decoder of claim 9 to 18.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011102435396A CN102419978B (en) | 2011-08-23 | 2011-08-23 | Audio decoder and frequency spectrum reconstructing method and device for audio decoding |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011102435396A CN102419978B (en) | 2011-08-23 | 2011-08-23 | Audio decoder and frequency spectrum reconstructing method and device for audio decoding |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102419978A CN102419978A (en) | 2012-04-18 |
| CN102419978B true CN102419978B (en) | 2013-03-27 |
Family
ID=45944360
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011102435396A Active CN102419978B (en) | 2011-08-23 | 2011-08-23 | Audio decoder and frequency spectrum reconstructing method and device for audio decoding |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102419978B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3107096A1 (en) | 2015-06-16 | 2016-12-21 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Downscaled decoding |
| CN111933162B (en) * | 2020-08-08 | 2024-03-26 | 北京百瑞互联技术股份有限公司 | Method for optimizing LC3 encoder residual error coding and noise estimation coding |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1677490A (en) * | 2004-04-01 | 2005-10-05 | 北京宫羽数字技术有限责任公司 | Intensified audio-frequency coding-decoding device and method |
| CN101253551A (en) * | 2005-08-30 | 2008-08-27 | Lg电子株式会社 | Method and apparatus for encoding and decoding an audio signal |
| CN101651457A (en) * | 2008-08-14 | 2010-02-17 | 杭州士兰微电子股份有限公司 | Audio coder and decoder, and coding and decoding methods |
| CN101743586A (en) * | 2007-06-11 | 2010-06-16 | 弗劳恩霍夫应用研究促进协会 | Audio encoder, encoding method, decoder, decoding method and encoded audio signal |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100539437C (en) * | 2005-07-29 | 2009-09-09 | 上海杰得微电子有限公司 | A kind of implementation method of audio codec |
| US8392176B2 (en) * | 2006-04-10 | 2013-03-05 | Qualcomm Incorporated | Processing of excitation in audio coding and decoding |
-
2011
- 2011-08-23 CN CN2011102435396A patent/CN102419978B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1677490A (en) * | 2004-04-01 | 2005-10-05 | 北京宫羽数字技术有限责任公司 | Intensified audio-frequency coding-decoding device and method |
| CN101253551A (en) * | 2005-08-30 | 2008-08-27 | Lg电子株式会社 | Method and apparatus for encoding and decoding an audio signal |
| CN101743586A (en) * | 2007-06-11 | 2010-06-16 | 弗劳恩霍夫应用研究促进协会 | Audio encoder, encoding method, decoder, decoding method and encoded audio signal |
| CN101651457A (en) * | 2008-08-14 | 2010-02-17 | 杭州士兰微电子股份有限公司 | Audio coder and decoder, and coding and decoding methods |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102419978A (en) | 2012-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101996636B (en) | Sub-band voice codec with multi-stage codebooks and redundant coding | |
| US7792668B2 (en) | Slot position coding for non-guided spatial audio coding | |
| RU2645271C2 (en) | Stereophonic code and decoder of audio signals | |
| CN101371447B (en) | Complex-transform channel coding with extended-band frequency coding | |
| KR101168473B1 (en) | Audio encoding system | |
| CN102047564B (en) | Factorization of overlapping transforms into two block transforms | |
| CN105074818A (en) | Methods for parametric multi-channel encoding | |
| CN101253809B (en) | Method and apparatus for encoding and decoding an audio signal | |
| CA2532288C (en) | Device and method for conversion into a transformed representation or for inversely converting the transformed representation | |
| CN101223573A (en) | Selectively using multiple entropy models in adaptive coding and decoding | |
| TW201214949A (en) | Adaptive hybrid transform for signal analysis and synthesis | |
| CN103544957A (en) | Method and device for bit distribution of sound signal | |
| Britanak et al. | Cosine-/sine-modulated filter banks | |
| CN101290774B (en) | Audio encoding and decoding system | |
| CN101185123B (en) | Scalable encoding device, and scalable encoding method | |
| CN115116451A (en) | Audio decoding method, audio encoding method, audio decoding device, audio encoding device, electronic equipment and storage medium | |
| CN101127212A (en) | Audio encoding | |
| CN101615911B (en) | A codec method and device | |
| CN102419978B (en) | Audio decoder and frequency spectrum reconstructing method and device for audio decoding | |
| KR20130015430A (en) | Method and apparatus for down-mixing multi-channel audio | |
| Shin et al. | Audio coding based on spectral recovery by convolutional neural network | |
| CN118571234A (en) | Audio encoding and decoding method and related device | |
| CN101741504B (en) | Method and device for determining linear predictive coding order of signal | |
| CN102592601B (en) | Signal processing method and device | |
| CN103489450A (en) | Wireless audio compression and decompression method based on time domain aliasing elimination and equipment thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20170306 Address after: Room 32, building 3205F, No. 707, Zhang Yang Road, free trade zone,, China (Shanghai) Patentee after: Xin Xin Finance Leasing Co.,Ltd. Address before: 201203 Shanghai city Zuchongzhi road Pudong New Area Zhangjiang hi tech park, Spreadtrum Center Building 1, Lane 2288 Patentee before: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20170712 Address after: 100033 room 2062, Wenstin Executive Apartment, 9 Financial Street, Beijing, Xicheng District Patentee after: Xin Xin finance leasing (Beijing) Co.,Ltd. Address before: Room 32, building 707, Zhang Yang Road, China (Shanghai) free trade zone, 3205F Patentee before: Xin Xin Finance Leasing Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20120418 Assignee: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. Assignor: Xin Xin finance leasing (Beijing) Co.,Ltd. Contract record no.: 2018990000163 Denomination of invention: Audio decoder and frequency spectrum reconstructing method and device for audio decoding Granted publication date: 20130327 License type: Exclusive License Record date: 20180626 |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20200306 Address after: 201203 Zuchongzhi Road, China (Shanghai) pilot Free Trade Zone, Pudong New Area, Shanghai 2288 Patentee after: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. Address before: 100033 room 2062, Wenstin administrative apartments, 9 Financial Street B, Xicheng District, Beijing. Patentee before: Xin Xin finance leasing (Beijing) Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200603 Address after: 361012 unit 05, 8 / F, building D, Xiamen international shipping center, No.97 Xiangyu Road, Xiamen area, China (Fujian) free trade zone, Xiamen City, Fujian Province Patentee after: Xinxin Finance Leasing (Xiamen) Co.,Ltd. Address before: 201203 Zuchongzhi Road, China (Shanghai) pilot Free Trade Zone, Pudong New Area, Shanghai 2288 Patentee before: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. Assignor: Xin Xin finance leasing (Beijing) Co.,Ltd. Contract record no.: 2018990000163 Date of cancellation: 20210301 |
|
| EC01 | Cancellation of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20120418 Assignee: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. Assignor: Xinxin Finance Leasing (Xiamen) Co.,Ltd. Contract record no.: X2021110000010 Denomination of invention: Audio decoder, audio decoding spectrum reconstruction method and device Granted publication date: 20130327 License type: Exclusive License Record date: 20210317 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230724 Address after: 201203 Shanghai city Zuchongzhi road Pudong New Area Zhangjiang hi tech park, Spreadtrum Center Building 1, Lane 2288 Patentee after: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. Address before: 361012 unit 05, 8 / F, building D, Xiamen international shipping center, 97 Xiangyu Road, Xiamen area, China (Fujian) pilot Free Trade Zone, Xiamen City, Fujian Province Patentee before: Xinxin Finance Leasing (Xiamen) Co.,Ltd. |
|
| TR01 | Transfer of patent right |