EP2589045A1 - Codage/décodage prédictif linéaire adaptatif - Google Patents
Codage/décodage prédictif linéaire adaptatifInfo
- Publication number
- EP2589045A1 EP2589045A1 EP11737984.2A EP11737984A EP2589045A1 EP 2589045 A1 EP2589045 A1 EP 2589045A1 EP 11737984 A EP11737984 A EP 11737984A EP 2589045 A1 EP2589045 A1 EP 2589045A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- filter
- current block
- modified
- signal
- coefficients
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000003044 adaptive effect Effects 0.000 title description 2
- 230000005236 sound signal Effects 0.000 claims abstract description 14
- 238000000034 method Methods 0.000 claims description 46
- 230000005540 biological transmission Effects 0.000 claims description 16
- 238000001914 filtration Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 2
- 230000005284 excitation Effects 0.000 description 15
- 238000004364 calculation method Methods 0.000 description 8
- 230000003595 spectral effect Effects 0.000 description 8
- 230000008901 benefit Effects 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000001228 spectrum Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 230000001755 vocal effect Effects 0.000 description 5
- 238000013459 approach Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000000295 complement effect Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 210000004072 lung Anatomy 0.000 description 1
- 229940050561 matrix product Drugs 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000003607 modifier Substances 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 210000001260 vocal cord Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/04—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using predictive techniques
- G10L19/06—Determination or coding of the spectral characteristics, e.g. of the short-term prediction coefficients
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/04—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using predictive techniques
- G10L19/16—Vocoder architecture
- G10L19/18—Vocoders using multiple modes
Definitions
- the object of the invention relates to the field of encoding / decoding audio and / or video data.
- the invention may relate to the coding of sounds with alternating speech and music.
- CELP Code-excited linear prediction
- CELP coders are predictive coders and are intended to model the production of speech from various elements such as:
- a stochastic excitation for example a white noise or an algebraic excitation
- a long-term prediction to model the vibration of the vocal cords, in period voiced in particular
- a short-term prediction in the form of a P-LPC (Linear Predictive Coding) filter, for modeling vocal tract modifications, such as the pronunciation of voiced consonants.
- P-LPC Linear Predictive Coding
- This number of coefficients P is chosen in order to correctly model the formational structure of the speech signal. Since the speech signal generally has four formants in the frequency band 0 to 4 kHz, ten filter coefficients correctly model this structure (two coefficients are necessary to model each formant).
- an LPC order of 16 coefficients is typically employed.
- the spectrum of a speech signal (in solid lines) is superimposed (in dotted lines) on the frequency response of an LPC filter modeling its spectral envelope.
- the power of the residual signal r can be small and its spectrum flat, by a judicious choice of the coefficients ⁇ ,.
- the residual signal is then easier to encode than the signal itself. It can be easily modeled by a strongly periodic harmonic signal, as shown in FIG. X (J) is the spectrum of the original signal s (black line) and E (/) is the spectrum of the residual signal r (in gray line).
- the coefficients a are typically calculated by a correlation measurement on the signal s "(and by applying a Levinson-Durbin type algorithm to invert the Wiener-Hopf equations).
- vocal tract modeling by the short-term prediction that models the spectral envelope in the form of a LPC filter
- transform coding may be added in cases where the sounds do not respond to the speech production model.
- CELP + TCX for "Transform Coded eXcitation”
- the quality of the coding according to the AMR WB + is satisfactory for the audio signals consisting of speech mixtures with background noise or speech with musical background, so typically for the signals where the speech dominates in energy.
- the envelope transmitted in LPC form is a relevant parameter since the signal consists mainly of speech which is described well thanks to a LPC envelope of a given order.
- the envelope indeed describes the formants (related to the resonance frequencies of the vocal tract) as a function of the number of coefficients chosen.
- the LPC envelope estimated and transmitted to the encoder is no longer sufficient.
- the audio signal is then often too complex to be limited, for example, to five formants, and its evolution over time makes a fixed number of coefficients is not adapted.
- the use of past information does not make it possible to anticipate the evolutions of the audio signal because using a backward predictor is relevant for a stationary signal but the spectrum at a given frame is precisely modeled and can be used to a next frame only if the statistical and especially spectral properties of the signal remain stable. Otherwise, the estimated LPC filter is irrelevant to the frame under consideration and the residual signal remains difficult to encode. The rear predictor loses all its interest.
- a solution recommended in the state of the art is therefore to use a switch between a prediction filter "before", calculated on the current frame, and a prediction filter, calculated on the previously received signal.
- the encoder analyzes the signal and decides whether the signal is stationary or not. If the signal is stationary, the back filter is used. Otherwise, a forward filter with few coefficients is transmitted to the decoder.
- Such an embodiment makes it possible to precisely control the quality of the signal residual to be encoded. It is implemented in the ITU-T G.729-E standard, in which a decision on stationarity of the signal results in an estimated "back” filter with 30 coefficients, or an estimated “forward” filter with 10 coefficients.
- the present invention improves the situation.
- a method of encoding a digital audio signal comprising a succession of consecutive blocks of data, from a predictive filter.
- the method within the meaning of the invention comprises in particular the use of a modified predictive filter for the coding of at least one current block.
- This modified filter is constructed by the combination of:
- the invention has many advantages: it makes it possible in particular not to switch abruptly from a rear filter to a front filter, but may for example offer the possibility of a transition by such a modified filter, in particular between the use of a rear filter and that of a front filter. It also avoids the passage through a filter before few coefficients to code a stationary signal with a complex envelope while it is only slightly disturbed by non-stationarity.
- Another advantage is to enrich a rear filter by producing an optimum coding quality without necessarily transmitting a complete front filter, in particular with as many coefficients as for example a front filter.
- Another advantage is to allow more choice to the encoder with different categories of filters: rear, front and modified.
- Enrichment parameters include the coefficients of a modifying filter, and the modified filter is constructed by a combination of back filter and modifier filter.
- This combination may be, in an exemplary embodiment described below, a convolution of the rear filter by the modifying filter. Alternatively, in another space, it may be a multiplication for example, or other.
- Such an embodiment has the advantage of allowing a simplification of the calculation operations with a decoder receiving the aforementioned parameters.
- the method may comprise, for the coding of a current block, a choice based on at least one predetermined criterion, a predictive filter of at least:
- a front filter adapted for the current block
- a modified filter estimated on the basis of a back filter and as a function of the signal in the current block.
- This criterion can for example take into account a stationarity of the signal between the past block and the current block, for the choice of one of the filters from a rear filter, a front filter and a modified filter.
- the predetermined criterion may comprise an estimate of a prediction gain based on a ratio between the power of the signal in the current block and the power of a residual signal after filtering this signal using each of the rear filters. , before and modified. Such an embodiment will be described in detail below, in particular with reference to FIGS. 4 and 5.
- the aforementioned criterion can also take into account a number of parameters to be sent to a decoder for the decoding of a current block and comprising at least coefficients that comprise the filter to choose.
- the predetermined criterion may include an optimum search between:
- the method then comprises the steps:
- c) calculating a plurality of modified filters of respective distinct orders, each estimated on the basis of a backward filter determined in step b) and as a function of the signal in a current block to be coded, d) comparing, for the same number of parameters to be sent to a decoder, this number being determined according to the filter orders, the performance of at least two filters among the front filters, the rear filters and the modified filters determined in steps a), b) and c ), and e) selecting, for the coding of a current block, a predictive filter having the best performance according to the comparison of step d), for a given number of parameters to be sent to a decoder.
- the modifying filter can be estimated by any technique, such as for example:
- the method may further comprise communication to a decoder of information of the type: choice of a front filter for a current block, with a transmission of parameters representing coefficients of the front filter,
- the present invention thus also aims at a method of decoding a digital audio signal comprising a succession of consecutive blocks of data, the method using a predictive filter for the decoding of a current block, the method comprising in particular:
- the decoding method may then comprise a step in which, for the decoding of at least one given current block, the predicted filter thus modified is used instead.
- this combination may consist of a multiplication or convolution (or other) of the rear filter by the modifying filter.
- the decoder may also use a rear filter or a front filter, according to the information received from the encoder.
- the back filter can be reconstructed based on previously decoded data. For example, it is possible to use the residual signal that the decoder has received from the encoder, for a past block, if the order of the rear filter to be reconstructed is higher than a previously constructed filter for this past block.
- the decoding method may thus comprise the steps, for the determination of the rear filter: determining a rear filter order, as a function of said received information, and estimating the rear filter, from previously decoded data and using this order of filtered.
- the "filter order" information may be transmitted directly from an encoder to the decoder, or may consist of implicit information.
- the decoder can be programmed to compute a NI coefficients back filter if a modified filter is to be constructed and calculate a N2 coefficients back filter for example if it is intended to use only a back filter for decoding.
- the invention proposes a combination of rear filter and a modifying filter chosen to complement each other and to create a modified filter of better quality than the rear filter, since it is a version of the rear filter, enriched. by an update resulting from the characteristics taken from the current block.
- the signal envelope is precisely described (for any type of signal), with an optimal transmission rate, whether in the form of a front filter, a rear filter or again a modified filter.
- the transition between filter is smooth compared to the prior art and thus avoids the discontinuity effect described above with reference to the prior art.
- the coding quality resulting from the use of the invention is then improved.
- Figure 3 schematically illustrates a succession of blocks signal in the form of a frame, for the choice of a relevant filter, in particular for coding the signal
- FIG. 4 illustrates an example of a prediction gain offered by the choice of a modified filter A 15 or a rear filter B 15 or a front filter Fi, depending on the order of this filter
- FIG. 5 illustrates an example of a prediction gain that a filter offers as a function of the bit rate requested by the choice of this filter, necessary for the transmission of its coefficients (or of its parameters of enrichment of the back filter to be transmitted, for example in the form of ISF indices for a modified filter A i5, as will be seen in an exemplary embodiment described below),
- FIG. 6A schematically illustrates an encoding device in one embodiment of the invention
- FIG. 6B schematically illustrates the steps of an encoding method in an embodiment of the invention
- FIG. 7A schematically illustrates a decoding device in one embodiment of the invention
- FIG. 7B schematically illustrates the steps of a decoding method in an embodiment of the invention.
- the notations used in the following are thus defined:
- LPC Linear Predictive Coding
- This technique can therefore be of the CELP type, for example according to the G.729, AMR or AMR-WB standards, or else a complementary coding transform can be used, for example in the sense of the G.718, G.729.1, AMR standards.
- WB +, MPEG-D "Unified Speech and Audio Coding").
- the filtering is intended to separate the signal to be coded into two components:
- the LPC filter A (z) is thus of the form:
- the number P denotes the number of non-zero coefficients. It is called "the order of the filter". Usually, a good number for a narrow-band speech signal (sampled at 8 kHz) is 10. This order can be increased, however, to better model the spectrum. signal and in particular to accentuate the precision of its envelope. It can also be increased if the signal sampling frequency is higher.
- the residual signal can also be presented in the perceptual weighted domain.
- a modification of this filter is used to better take into account the properties of the human ear during the coding of the residual.
- the coefficients a t of the LPC filter are commonly estimated by identifying the audio signal and its least squares prediction. We therefore seek the coefficients a t minimizing the quadratic error of the passed audio signal, through the filter A (z). It is therefore sought to minimize the power of the signal r n . This power is estimated over a certain period representing a number of samples N. The coefficients are therefore valid for this period of time.
- This estimation of LPC filter coefficients is thus carried out by estimating the autocorrelation terms of the signal x n , and by solving the equations of Yule Walker or Wiener Hopf, typically by a fast algorithm of the Levinson Durbin type, as described by example in the reference:
- the estimation of the coefficients of the LPC filter can be performed on the current signal x n , on a frame representing a set of samples, or on a version of the signal x m (m ⁇ n) resulting from a previous local decoding ( complete or partial) of the signal in coded form.
- Local decoding is obtained by decoding the encoded parameters at the encoder. This local decoding makes it possible to recover, at the level of the coder, the information that can be used by the decoder in the same way.
- the LPC filter is calculated on the original samples of the current frame (t-frame), or previous frames (t-1, t-2, etc.): in such cases, it is a LPC filter "Before” and its coefficients (hereinafter referred to as ”) shall be communicated to the decoder, or
- the LPC filter can be calculated from locally decoded samples, thus older than the current frame (t-1, t-2, etc.): in this case, it is a "backward" LPC filter and the decoder is also capable of estimating the coefficients (denoted b n ) of the same LPC filter, which therefore does not need to be communicated to the decoder.
- the performance of the LPC filter can then be evaluated by estimating the power of the residual signal (i.e. the signal strength resulting from the filtering of the original signal of the current frame by the LPC filter considered).
- the ratio of the power of the original signal divided by the power of the residual signal gives a quantity called "prediction gain", often expressed in dB.
- the LPC filters are estimated in forward mode, on the current frame, and in backward mode on the decoded previous frame. Their own prediction gain is then calculated.
- the gain of the front LPC filter is always better than the gain of the rear LPC filter for a given order.
- the rear LPC filter is not adapted to treat the current frame, but rather the previous frame.
- it often happens (as in the case presented here as an example), in particular when the signal is stationary, that the gain of a rear LPC filter is greater than the prediction gain of a front LPC filter. lower order.
- the prediction gain is greater in the backward mode with an order 24 than in the forward mode with an order of 10 or 16.
- the filter flO requires the transmission of its coefficients to the decoder, while the filter b24 is computable to the decoder without the need to transmit additional information. Nevertheless, the filter b24 has a much lower prediction gain than the prediction gain of the filter f24 (before filter of the same length).
- This filter A hereinafter called “modified filter” is then used by the (possibly weighted) coder to calculate the residue.
- An inverted version (1 / A (z)) of this filter is used at the decoder to reshape the spectrum of the signal.
- the modifying filter can be calculated conventionally by the Levinson Durbin algorithm acting on the signal derived from the filtering of the signal of the current frame by the determined back filter.
- the modifying filter can be determined on the basis of an analysis of a residual signal obtained after filtering the current block by a back filter calculated for a past block.
- the modifying filter (M) can be estimated by "deconvolution".
- the modifying filter can be estimated, according to this first option, by deconvolving a front filter adapted for filtering the current block, by a back filter calculated for a past block.
- the modifying filter can be estimated by a least squares Wiener identification method in which the autocorrelation terms of the filter are calculated. rear (r 0 , r hr J ), and the cross-correlation between the target front filter and the back filter ( ⁇ 3 ⁇ 4, ⁇ 3 ⁇ 4 ... Cq.j), the filter M then being obtained by the matrix product next: m 0 1 1 q-2 1 ⁇ ql c o
- this second option can be implemented by least squares identification, by calculating autocorrelation terms of coefficients of the back filter and intercorrelation between the modified filter and the back filter.
- the second option can be executed in practice by a fast algorithm (of the type used for the identification of the LPC coefficients and based on the autocorrelation of the signal). Nevertheless, the first option of deconvolution can be as advantageous.
- the filter M obtained by any of these techniques is then typically quantified in a form suitable for the transmission of the LPC filter coefficients (for example using a conversion of the LSF, LSP or ISF type (for "line spectral frequencies", or “peers”)). Once quantized, these coefficients are convoluted with the rear filter B to obtain a filter A (z) which can be reproduced identically to the decoder.
- the performances of the obtained filter are compared with those of the quantized before filter (F) containing the same number of coefficients as the calculated filter M. If the number of bits used to transmit a filter depends only on the length of the filter (which is often the case in speech / audio coding), then the performances between the filter A and the filter F can be directly compared by their prediction gain, calculated on the original signal x n . So :
- the filter M is transmitted, and, if not, the filter F is transmitted.
- the filter A is of a higher order than the filter F (thus rendering its estimate expensive for the decoder since it involves the estimation of the filter B and the decoding of the filter M), the filter A is selected only if its gain of prediction is much higher than the filter F (a few dB).
- An embodiment presented hereinafter therefore considers the calculation of a plurality of rear filters, before and modifying.
- the number of front filters is not necessarily the same as the number of back filters.
- a set of back filters determined is calculated, according to the method presented above, a set of quantized modifying filters. It is advisable to choose modifying filters with orders identical to the orders of the filters before F already calculated (pf 0 , pfi, pf 2 , pf 3 ) -
- the convolution of the rear filters (B) and the modified filters (M) then gives a set of combined filters (A) whose performances are compared with those of the front filters (in particular those of the front filters having an order identical to the modified filter M).
- FIG. 4 shows the performances of the back filters calculated at 5 different orders (from B 0 of order pbo to B 4 of order pb 4 ). It is observed that the filter B 4 has poorer performance than the filter B 3 . This filter, like any rear filter performance lower than a rear filter of lower order, is immediately removed from later considerations. This avoids unnecessarily calculating modified filters based on this filter B 4 .
- the performances of the filters calculated beforehand with 4 different orders (of F 0 of order pf 0 to F 3 of order pf 3 ) are also represented.
- the abscissa of the graph of Figure 4 represents the order of prediction and the ordinate, the prediction gain.
- a filter modifying (Mi ; 0 ) of order pf 0 is calculated to obtain a first filter A 0 .
- a filter modifying (M 2> 0 ) of order pf 0 is calculated to obtain a second filter
- a filter modifying (M 3> 0 ) of order pf 0 is calculated to obtain a third filter A 2 .
- a filter modifying (M 3> i) of order pfi is calculated to obtain a fourth filter A 3 .
- the filters A 0 , A 1 and A 2 therefore have an identical transmission cost because they require the transport of pfo coefficients. This transmission cost can be considered identical to that of the filter F 0 .
- the transmission cost of the filter A 3 is comparable to the transmission cost of the filter Fj.
- the filters By positioning the filters in the rate / gain coding scheme (FIG. 5), the best possibilities for coding the LPC envelope are finally selected. It appears that the relevant configurations are then the filters B 3 , A 0 or A 2 , F ⁇ , F 2 and F 3 . Other configurations, offering lower performance for the same or higher throughput, can be eliminated.
- the choice of the filter A 0 requires the calculation of the filter B ⁇ and the decoding of a filter modifying order pf 0
- the choice of the filter A 2 requires the calculation of the filter B 2 and the decoding of a filter modifying order pf 0 : this choice therefore implies more complexity than that of the filter A 0 for identical performance
- the choice of the filter B 3 requires the calculation of a high order filter pb 3 and therefore has more complexity.
- d represents the number of bits allocated to the transmission of the residue. This number can be estimated, knowing the total bit rate, for the coding of the audio frame (T), the number of samples it comprises (N) and the bit rate required for the coding of the LPC (R) filter, as follows:
- the filter of index 2 If this quantity is positive, one will choose the filter of index 2 (if not, the filter of index 1).
- the type of front / rear / combined filter can change from one frame to the next, depending on the choice made at the encoder. However, care should be taken to avoid configuration changes which are too fast if the prediction gains are not sufficiently different, in particular between the configuration used on the previous frame and the configuration giving the best performance on the current frame.
- a change is useful only beyond a certain threshold (for example 1 dB).
- the encoder must inform the decoder so that it can calculate the chosen LPC filter.
- Useful information for this purpose is for example:
- read index__pf 1 filter order before or read f [pf] ISF ... number of bits, depends on ⁇
- the filter coefficients are assumed to be quantized in their ISF form. They are grouped together to be coded together.
- a typical configuration used in the AMR-WB (3GPP) encoder is repeated in this embodiment. It is 46 bits for 16 LPC coefficients represented as ISF. For 10 coefficients, we will use 18 bits for example.
- Reading the 2-bit index_pb flag is associated with a corresponding number of filter coefficients. For example, the following association can be provided:
- index_pf flag can be represented on a single bit:
- the coefficients f n are interpreted as the coefficients of the filter modifying the rear filter. Otherwise the coefficients f n are interpreted as front filter coefficients.
- the syntax presented above can be arranged, or even simplified, if we reduce the number of combinations.
- the index_pb field can be omitted if only one possible back filter order is considered.
- the order of the rear filter can be implicitly set to 16.
- a single length can be envisaged, for example 16.
- the decoder On decoding, the decoder, on reading the information indicating the use of the rear filter and its order, calculates the rear filter of the order indicated on the decoded samples beforehand.
- the decoder On receipt of the presence indication and the order of a filter, it decodes the transmitted ISF indices to convert the filter into LPC filter coefficients. Of course, here, if only the back filter is signaled (without ISF indexes), the decoder understands that the filter used is finally only the rear filter (B). If the two filters are transmitted (with the ISF indices), the decoder understands that the filter A used is the "modified" filter (obtained by the convolution of the front and back filters (B * M), the filter M being interpreted as the modifying filter).
- the decoder understands that the filter used is the front filter alone.
- the present invention provides an alternative to the coding of the LPC envelope, a critical element for the quality of coding, especially in audio coding.
- an alternative mode of coding the LPC envelope does not cause any difficulty compared to current techniques: the encoder can always choose the standard mode LPC forward, as a fallback position.
- the decoder is able to use rear filters, especially when the signal is stationary. Nevertheless, he is also able to take advantage of both approaches by combining them.
- the performance of the LPC filter is further increased by increasing its accuracy to produce improved quality.
- the present invention also provides a device for encoding a signal for implementing the above coding method.
- An exemplary embodiment is shown in FIG. 6A and such an encoder D1 comprises, for example:
- CALC means for calculating a modified filter A on the basis of a rear filter and at least a function of the signal in the current block SGN-Tn (in a current frame Tn for example), and
- the encoder device determines a prediction gain Gp for a given bit rate d, by considering several types of filters before F, rear B and modified A and retains in step 12 the filter having for example the best prediction gain at this given rate d.
- the best candidate filter is a modified filter A (step 13)
- the construction thereof implies a filter modifying Mj, the order j of this modifying filter being able to be chosen according to the order i of the rear filter Bi on the base of which is built the modified filter A.
- the coefficients of the filter modifying Mj and the order i of the filter Bi can then be sent to a decoder device D2.
- the present invention also relates to a computer program comprising instructions for the implementation of these steps, when this program is executed by a processor, for example of such an encoding device Dl.
- a processor for example of such an encoding device Dl.
- the flowchart shown in Figure 6B may illustrate the general algorithm of such a program.
- the present invention also relates to the decoding device D2 of an encoded signal for implementing the decoding method.
- a device comprises at least: REC information receiving means (for example information representing the coefficients of the filter modifying Mj (in the form of ISF for example) and the order i of the filter rear Bi), for the calculation of a modified predictive filter A,
- the decoder device receives in step 20 information (for example from the coder D1), this information may comprise here:
- this rear filter Bi is calculated from previously decoded data (for example from a previous frame ⁇ ⁇ _ ⁇ ) and using the filter order i.
- the modifying filter Mj and the rear filter Bi thus calculated are combined (for example by convolution) to obtain, in step 23, the modified filter A used for decoding the signal by the decoder device D2.
- step 24 for a current frame to be delivered.
- the present invention also relates to a computer program comprising instructions for the implementation of these steps, when this program is executed by a processor, for example such a decoding device D2.
- a processor for example such a decoding device D2.
- FIG. 7B can illustrate the general algorithm of such a program.
- the program for implementing the encoding method (FIG. 6B) and the program for implementing the decoding method (FIG. 7B) can be grouped together in the same general computer program within the meaning of the invention .
- the present invention is not limited to the embodiment described above by way of example; it extends to other variants.
- the criterion for choosing a filter illustrated in FIG. 5 may not be simply limited to the best prediction gain for a given flow rate.
- another criterion that may be taken into consideration may be the complexity of calculations to the encoder or decoder.
- the modified filters A 0 and A 2 are the best candidates for the flow rate d 0 . It will then be chosen preferentially the filter A 0 , less complex than the filter A 2 , and nevertheless offering the same performance in terms of prediction gain.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Quality & Reliability (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR1055206A FR2961937A1 (fr) | 2010-06-29 | 2010-06-29 | Codage/decodage predictif lineaire adaptatif |
PCT/FR2011/051393 WO2012001260A1 (fr) | 2010-06-29 | 2011-06-17 | Codage/decodage predictif lineaire adaptatif |
Publications (2)
Publication Number | Publication Date |
---|---|
EP2589045A1 true EP2589045A1 (fr) | 2013-05-08 |
EP2589045B1 EP2589045B1 (fr) | 2014-04-16 |
Family
ID=42989275
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP11737984.2A Active EP2589045B1 (fr) | 2010-06-29 | 2011-06-17 | Codage/décodage prédictif linéaire adaptatif |
Country Status (4)
Country | Link |
---|---|
US (1) | US9620139B2 (fr) |
EP (1) | EP2589045B1 (fr) |
FR (1) | FR2961937A1 (fr) |
WO (1) | WO2012001260A1 (fr) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9583115B2 (en) * | 2014-06-26 | 2017-02-28 | Qualcomm Incorporated | Temporal gain adjustment based on high-band signal characteristic |
EP3270376B1 (fr) * | 2015-04-13 | 2020-03-18 | Nippon Telegraph and Telephone Corporation | Codage prédictif linéaire d'un signal sonore |
Family Cites Families (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4328585A (en) * | 1980-04-02 | 1982-05-04 | Signatron, Inc. | Fast adapting fading channel equalizer |
US5533052A (en) * | 1993-10-15 | 1996-07-02 | Comsat Corporation | Adaptive predictive coding with transform domain quantization based on block size adaptation, backward adaptive power gain control, split bit-allocation and zero input response compensation |
JP3064947B2 (ja) * | 1997-03-26 | 2000-07-12 | 日本電気株式会社 | 音声・楽音符号化及び復号化装置 |
FR2762464B1 (fr) * | 1997-04-16 | 1999-06-25 | France Telecom | Procede et dispositif de codage d'un signal audiofrequence par analyse lpc "avant" et "arriere" |
ATE302991T1 (de) * | 1998-01-22 | 2005-09-15 | Deutsche Telekom Ag | Verfahren zur signalgesteuerten schaltung zwischen verschiedenen audiokodierungssystemen |
US7072832B1 (en) * | 1998-08-24 | 2006-07-04 | Mindspeed Technologies, Inc. | System for speech encoding having an adaptive encoding arrangement |
US6449590B1 (en) * | 1998-08-24 | 2002-09-10 | Conexant Systems, Inc. | Speech encoder using warping in long term preprocessing |
US6456964B2 (en) * | 1998-12-21 | 2002-09-24 | Qualcomm, Incorporated | Encoding of periodic speech using prototype waveforms |
US7302387B2 (en) * | 2002-06-04 | 2007-11-27 | Texas Instruments Incorporated | Modification of fixed codebook search in G.729 Annex E audio coding |
DE102004025471A1 (de) * | 2004-05-21 | 2005-12-15 | Micronas Gmbh | Verfahren bzw. adaptives Filter zum Verarbeiten einer Folge aus Eingabe-Daten eines Funksystems |
KR101393298B1 (ko) * | 2006-07-08 | 2014-05-12 | 삼성전자주식회사 | 적응적 부호화/복호화 방법 및 장치 |
EP1883067A1 (fr) * | 2006-07-24 | 2008-01-30 | Deutsche Thomson-Brandt Gmbh | Méthode et appareil pour l'encodage sans perte d'un signal source, utilisant un flux de données encodées avec pertes et un flux de données d'extension sans perte. |
JP4299323B2 (ja) * | 2006-08-04 | 2009-07-22 | 株式会社日立国際電気 | 通信システム |
EP2054876B1 (fr) * | 2006-08-15 | 2011-10-26 | Broadcom Corporation | Dissimulation de perte de paquets pour codage predictif de sous-bande a base d'extrapolation de guide d'ondes audio pleine bande |
-
2010
- 2010-06-29 FR FR1055206A patent/FR2961937A1/fr active Pending
-
2011
- 2011-06-17 EP EP11737984.2A patent/EP2589045B1/fr active Active
- 2011-06-17 US US13/807,657 patent/US9620139B2/en active Active
- 2011-06-17 WO PCT/FR2011/051393 patent/WO2012001260A1/fr active Application Filing
Also Published As
Publication number | Publication date |
---|---|
FR2961937A1 (fr) | 2011-12-30 |
US20130103408A1 (en) | 2013-04-25 |
WO2012001260A1 (fr) | 2012-01-05 |
EP2589045B1 (fr) | 2014-04-16 |
US9620139B2 (en) | 2017-04-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1994531B1 (fr) | Codage ou decodage perfectionnes d'un signal audionumerique, en technique celp | |
TWI413107B (zh) | 具有多重階段編碼簿及冗餘編碼之子頻帶語音編碼/解碼的方法 | |
JP2020109529A (ja) | 復号方法、コンピュータプログラム及び復号システム | |
JP6692948B2 (ja) | 異なるサンプリングレートを有するフレーム間の移行による音声信号の線形予測符号化および復号のための方法、符号器および復号器 | |
EP1692689B1 (fr) | Procede de codage multiple optimise | |
EP1997103B1 (fr) | Procede de codage d'un signal audio source, dispositif de codage, procede et dispositif de decodage, signal, produits programme d'ordinateur correspondants | |
EP3175444B1 (fr) | Gestion de la perte de trame dans un contexte de transition fd/lpd | |
EP2080194B1 (fr) | Attenuation du survoisement, notamment pour la generation d'une excitation aupres d'un decodeur, en absence d'information | |
EP3175443B1 (fr) | Détermination d'un budget de codage d'une trame de transition lpd/fd | |
EP2795618B1 (fr) | Procédé de détection d'une bande de fréquence prédéterminée dans un signal de données audio, dispositif de détection et programme d'ordinateur correspondant | |
EP1875465A1 (fr) | Procede d'adaptation pour une interoperabilite entre modeles de correlation a cout terme de signaux numeriques | |
FR2880724A1 (fr) | Procede et dispositif de codage optimise entre deux modeles de prediction a long terme | |
EP2345029B1 (fr) | Procédé, programme informatique et entité de décodage d'un signal audio numérique | |
EP2589045B1 (fr) | Codage/décodage prédictif linéaire adaptatif | |
KR102099293B1 (ko) | 오디오 인코더 및 오디오 신호를 인코딩하는 방법 | |
Tosun | Dynamically adding redundancy for improved error concealment in packet voice coding | |
WO2013135997A1 (fr) | Modification des caractéristiques spectrales d'un filtre de prédiction linéaire d'un signal audionumérique représenté par ses coefficients lsf ou isf | |
FR2737360A1 (fr) | Procedes de codage et de decodage de signaux audiofrequence, codeur et decodeur pour la mise en oeuvre de tels procedes | |
FR2980620A1 (fr) | Traitement d'amelioration de la qualite des signaux audiofrequences decodes | |
FR3018942A1 (fr) | Estimation d'un bruit de codage introduit par un codage en compression de type micda |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 20121227 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
RAP1 | Party data changed (applicant data changed or rights of an application transferred) |
Owner name: ORANGE |
|
DAX | Request for extension of the european patent (deleted) | ||
GRAP | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOSNIGR1 |
|
RIC1 | Information provided on ipc code assigned before grant |
Ipc: G10L 19/06 20130101AFI20131010BHEP Ipc: G10L 19/18 20130101ALN20131010BHEP |
|
INTG | Intention to grant announced |
Effective date: 20131113 |
|
GRAS | Grant fee paid |
Free format text: ORIGINAL CODE: EPIDOSNIGR3 |
|
GRAA | (expected) grant |
Free format text: ORIGINAL CODE: 0009210 |
|
AK | Designated contracting states |
Kind code of ref document: B1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
REG | Reference to a national code |
Ref country code: GB Ref legal event code: FG4D Free format text: NOT ENGLISH |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: EP |
|
REG | Reference to a national code |
Ref country code: AT Ref legal event code: REF Ref document number: 662968 Country of ref document: AT Kind code of ref document: T Effective date: 20140515 |
|
REG | Reference to a national code |
Ref country code: IE Ref legal event code: FG4D Free format text: LANGUAGE OF EP DOCUMENT: FRENCH |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R096 Ref document number: 602011006255 Country of ref document: DE Effective date: 20140528 |
|
REG | Reference to a national code |
Ref country code: AT Ref legal event code: MK05 Ref document number: 662968 Country of ref document: AT Kind code of ref document: T Effective date: 20140416 |
|
REG | Reference to a national code |
Ref country code: NL Ref legal event code: VDEP Effective date: 20140416 |
|
REG | Reference to a national code |
Ref country code: LT Ref legal event code: MG4D |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: NO Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140716 Ref country code: IS Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140816 Ref country code: BG Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140716 Ref country code: CY Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: GR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140717 Ref country code: NL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: LT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: FI Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: HR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: PL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: LV Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: SE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: AT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: RS Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: ES Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: PT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140818 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R097 Ref document number: 602011006255 Country of ref document: DE |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: CZ Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: LU Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140617 Ref country code: RO Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: MC Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: DK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: SK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: EE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: PL |
|
PLBE | No opposition filed within time limit |
Free format text: ORIGINAL CODE: 0009261 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT |
|
26N | No opposition filed |
Effective date: 20150119 |
|
REG | Reference to a national code |
Ref country code: IE Ref legal event code: MM4A |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: IT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: IE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20140617 Ref country code: LI Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20140630 Ref country code: CH Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20140630 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R097 Ref document number: 602011006255 Country of ref document: DE Effective date: 20150119 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 5 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: SI Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: MT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: SM Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 6 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: TR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 Ref country code: BE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140630 Ref country code: HU Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT; INVALID AB INITIO Effective date: 20110617 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 7 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 8 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: MK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: AL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20140416 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: GB Payment date: 20240521 Year of fee payment: 14 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: DE Payment date: 20240521 Year of fee payment: 14 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: FR Payment date: 20240522 Year of fee payment: 14 |