Advanced processing based on a complex-exponential-modulated filterbank and adaptive time signalling methods

TECHNICAL FIELD [0001] The present invention relates to audio source coding systems but the same methods can also be used in many other technical fields. Various techniques are presented that are useful in audio coding systems using parametric representations.

BACKGROUND OF THE INVENTION AND BACKGROUND ART [0002] The present invention relates to parametric coding of a stereo image of an audio signal. Typical parameters used to describe stereo image properties are inter-channel intensity difference (IID), inter-channel time difference (ITD) and inter-channel coherence (IC). In order to reconstruct the stereo image based on these parameters, a method is needed that allows reconstruction of the correct level of correlation between the two channels according to the IC parameter. This is achieved by means of a de-correlation method.

There are several methods available for creating de-correlated signals, such as the method described in US6005946. In an ideal situation, a linear time invariant (LTI) function with a high-frequency response is desirable. One obvious way to achieve this is to use a fixed delay. However, the use of delay or any other full-band LTI function will trigger a low-band response when the raw signal is added. In the event of a delay, the result will be a typical comb filter. The comb filter often produces unwanted "metallic" sound, which - even if the stereo widening effect is efficient - significantly reduces the naturalness of the original.

[0003] Methods in the frequency domain for generating a signal correlated by adding a random sequence to an IDD value along a frequency axis in which different sequences for different audio channels are used are also known in the art. One of the problems with decorrelation in the frequency domain by modifying random sequences is to introduce leading echoes. Subjective tests have shown that for non-stationary signals, leading echoes are much more annoying than ordinary echoes, which is well justified by the established principles of psychoacoustics. This problem could be reduced by dynamically adapting the transform size to the signal properties in terms of transient content. However, resizing transforms is always a difficult decision (i.e. binary) that affects the entire signal bandwidth and which can be difficult to implement reliably.

[0004] In US Patent Application US 2003/0219130 A1, coherence-based coding and synthesis are disclosed. In particular, the sound stage is synthesized from a monophonic audio signal by modifying, for each critical band, an audio stage parameter such as inter-aural level difference (ILD) and / or inter-aural time (ITD) difference) for each subband in the critical band, the modification being based on the estimated average coherence for the critical band. Modification based on coherence produces sound scenes with the width of objects that more accurately match the width of the objects in the original input sound stage. Stereo parameters are well-known BCC parameters, where BCC stands for Binaural Cue Coding. When generating two different de-correlated output signals, the frequency coefficients obtained by the discrete Fourier transform are grouped together in one critical band. Based on the measure of inter-channel coherence, the weighting factors are multiplied by a pseudo-random sequence, which is preferably chosen in such a way that the variance is approximately constant for all critical bands and the mean in each critical band is "0". The same sequence is used for the spectral coefficients of each other frame.

[0005] WO 03/007656 A1 discloses an apparatus and method for generating audio output signals having a specific inter-correlation relationship. The device works by shifting the input signal frequency bands in phase by deferring values ​​that depend on the desired mutual correlation. The input signal's amplitude spectrum does not change.

Summary of the Invention [0006] The object of the present invention is to provide a device and method for generating a de-correlation signal.

[0007] This object is achieved by means of a device for generating a de-correlation signal according to claim 1, a multi-channel decoder according to claim 8, a method for generating a de-correlation signal according to claim 1. 13, a multi-channel decoding method according to claim 1. Or a computer program according to claim 15.

[0008] The present invention is based on the finding that a well-correlated signal is generated on the decoding side for generating the first and second multi-channel signal channels based on the mono input signal when a reverberation filter is used which introduces an overall or preferably fractional delay to the input signal . Importantly, this reverberation filter is not applied to the entire input signal. Instead, several reverberation filters are used for several subbands of the original input signal, i.e. mono signal, so that reverberation filtering using reverberation filters is not applied in the time domain or in the frequency domain, i.e. in the field obtained using Fourier transform. According to the invention, the reverberation filtering using the reverberation filters for subbands is carried out individually in the subband domain.

[0009] The subband signal comprises a sequence of at least two subband samples, the subband sample sequence representing the subband signal bandwidth that is smaller than the input bandwidth. Naturally, the bandwidth of the subband signal is greater than the bandwidth assigned to the frequency coefficients obtained by the Fourier transform. The subband signals are preferably generated by a filter bank having, for example, 32 or 64 filter bank channels, while the FFT transformation for the same example will contain 1024 or 2048 frequency coefficients, i.e. frequency channels.

[0010] The subband signals may be subband signals obtained by subband filtering of the input signal sample block. Alternatively, the subband filter bank can also be used continuously without block processing. However, block processing is preferred in the present invention.

[0011] Because reverberation filtering is not applied to the entire signal but is used subband, the "metallic" sound caused by comb filtering is avoided.

[0012] In cases where the sample period between two consecutive subband samples is too large to provide a good listening experience on the decoder side, it is preferable to use fractional delays in the reverberation filter, e.g. a delay of 0.1 to 0.9, and preferably 0.2 to 0.8, the sampling period of the subband signal. It should be noted that for critical sampling and when 64 subband signals are generated using a filter bank having 64 filter bank channels, the sampling period in the subband signal is 64 times the sampling period of the original input signal.

[0013] It should be noted that in this case delays are an integral part of the filtering process used in the reverberation device. The output signal consists of many delayed versions of the input signal. It is advantageous to delay the signals by a fraction of the subband sampling period to obtain a good reverberation device in the subband domain.

[0014] In preferred embodiments of the present invention, the delay and preferably the fractional delay introduced by each reverberation filter in each subband is the same for all subbands. Regardless, the filter coefficients are different for each subband. The use of IIR filters is beneficial. Depending on the actual situation, the fractional delay and filter coefficients for different filters can be determined experimentally using listening tests.

[0015] The subbands filtered by means of a set of reverberation filters create a de-correlation signal to be mixed with the original input signal, i.e. mono signal to obtain the decoded left channel and the decoded right channel. This mixing of the decorrelation signal with the original signal is implemented based on the inter-channel coherence parameter sent together with the parametrically coded signal. In order to get different left and right channels, i.e. different first and second channels, mixing the de-correlation signal with the mono signal to obtain the first output channel is different from mixing the de-correlated signal with the mono signal to obtain the second output channel.

[0016] For greater coding efficiency, multi-channel coding is performed using adaptive stereo parameter set. To this end, the encoder includes, in addition to the means for calculating the mono signal and in addition to the means for generating a set of stereo parameters, means for determining the validity of the set of stereo parameters for the subsequent parts of the left and right channels. Preferably, the determining means operate to activate the generating means when it is determined that the stereo parameter set is no longer valid, so that the second parameter set is calculated for the left and right channel parts starting from the second time limit. The second time limit is also set by means of determining validity.

[0017] Thus, the encoded output signal comprises a mono signal, a first set of stereo parameters and a first time limit associated with the first set of parameters and a second set of parameters and a second time limit associated with the second set of stereo parameters. On the decoder side, the decoder will use a valid set of stereo parameters until it reaches a new time limit. When the new time limit is reached, decoding operations are performed using the new stereo parameter set.

[0018] Compared with prior art methods in which block processing took place and hence block determination of the stereo parameter set, the adaptive determination of the stereo parameter set according to the invention for different set time limits on the decoder side ensures high coding efficiency on one hand and high quality coding on the other hand. This is because for relatively stable signals, the same set of stereo parameters can be used for multiple blocks of mono signal samples without introducing audible errors. On the other hand, when dealing with non-stationary signals, the adaptive stereo parameterization according to the invention ensures better time resolution so that each part of the signal has its optimal set of stereo parameters.

[0019] The present invention offers a solution to the problems of the prior art by using the reverberation unit as a decorrelator implemented with fractional delay lines in the filter bank and using adaptive level of the de-correlated reverberated signal.

[0020] Several aspects of the present invention are outlined below.

[0021] One aspect of the invention is a method of delaying a signal by: filtering a time domain real value signal through part of the analysis of a complex filter bank; modification of subband signals with complex values ​​by part of the filter bank synthesis; and the use of the real part of the output signal with complex values ​​in the time domain, the output signal being the sum of the signals obtained from the synthesis filtering.

[0022] Another aspect of the invention is a method of modifying complex values ​​subband signals by filtering each complex signal subband with a complex values ​​filter with a finite impulse response, wherein the finite impulse response filter for the n number of the subband is given by Fourier transform of exp (-i / y (in +1 / 2) r) G_f (<W)_t for even n exp (-j; r (n +] / 2) r) C?_r(<w + / r), for odd n where parameter τ = T / L, and in which the synthesis filter bank has L subbands and the desired delay is T measured in units of the output signal samples.

[0023] Another aspect of the invention is a method of modifying complex band signals by filtering, in which the filter Gro) approximately satisfies the relationship / <o) GTo) + / (o + n) GTo + n) = 1, where / To) is a discrete time transformation ν, (ί) = ^ ί * Σρ (ΟΡ (ί-Γ-Ι *),

Fourier ap () sequence is a prototype filter of the mentioned complex value filter bank, and A is a proper normalization factor with real values.

[0024] Another aspect of the invention is a method of modifying subband signals with complex values ​​by means of filtering in which the filter GTo) satisfies the relationship GT-o) = GTo + n) *, so that the impulse response samples with even indexes have real values ​​and the response samples impulse with odd indexes have only imaginary values.

[0025] Another aspect of the invention is a method of coding the stereo properties of the input signal, on the encoder side, by calculating time grid parameters describing the location in time for each set of stereo parameters, where the number of stereo parameters is arbitrary, and on the decoder side, by using parametric stereo synthesis according to the time grid.

[0026] Another aspect of the invention is a method of coding the stereo properties of an input signal in which the time location of the first stereo parameter set, in which the time hand for the stereo parameter set coincides with the beginning of the frame, is signaled directly instead of time indicator signaling.

[0027] Another aspect of the invention is a method of generating stereo de-correlation for parametric stereo reconstruction by, on the decoder side, using an artificial reverberation process for the synthesis of an additional signal.

[0028] Another aspect of the invention is a method of generating stereo de-correlation for parametric stereo reconstruction by, on the decoder side, a reverberation process that is performed in a complex modulated filter assembly using phase delay matching on each channel of the filter assembly.

Another aspect of the invention is a method of generating stereo de-correlation for parametric stereo reconstruction in that, on the decoder side, the reverberation process uses a detector designed to find signals in which the reverberation tail could be undesirable and allows the reverberation tail to be suppressed or removed.

BRIEF DESCRIPTION OF THE DRAWINGS [0029] The present invention will now be described by way of illustrative examples that do not limit the scope of the invention, with reference to the accompanying drawings in which:

Fig. 1 illustrates a block diagram of an apparatus according to the invention;

Fig. 2 illustrates a block diagram of a means for generating a de-correlated signal;

Fig. 3 illustrates single channel analysis and synthesis of a pair of stereo channels based on the reconstructed subband stereo signals of the present invention;

Fig. 4 illustrates a block diagram of the division of parametric sets of stereo parameters into time segments based on signal properties; and

Fig. 5 illustrates an example of the division of parametric sets of stereo parameters in time segments based on signal properties.

DESCRIPTION OF PREFERRED EMBODIMENTS [0030] The embodiments described below only illustrate the principles of the present invention of parametric stereo coding. It is understood that modifications and changes to the arrangements and details described herein are obvious to those skilled in the art. It is therefore intended that the invention be limited only by the scope of the following claims, and not by the particular details presented by way of description and explanation of the present embodiments.

[0031] Signal delay by a fraction of a sample can be achieved by several prior art interpolation methods. However, there are special cases when the original signal is available as oversamplified samples with complex values. The implementation of the fractional delay in the QMF filter bank only by applying a phase delay by a factor, for each QMF channel corresponding to a fixed time delay, brings severe artifacts.

[0032] This can be efficiently avoided by using a compensating filter according to an innovative approach enabling high-quality approximation of arbitrary delays in any filter bank exponentially modulated in complex values. A detailed description is given below.

Continuous time model [0033] To facilitate calculations, the L-band filter bank exponentially modulated in complex values ​​will be modeled here by continuous-windowed transformation using synthesis waveforms

<img file="PL2265041T3_D0001.tif" />

(1) where n, k are integers with n&gt; 0 and θ is a member of the solid phase. The results for discrete time signals are obtained by appropriate sampling of the variable t at 1 / L intervals. It is assumed that the v (t) window with real values ​​is selected so that for x (t) signals with real values ​​with high accuracy the relationship is true

<img file="PL2265041T3_D0002.tif" />

(2) if ^ (*) = (3) where * is a complex join. It is also assumed that v (t) is essentially band limited to the frequency range [-π, π]. Let's consider the modification of each n-frequency band by filtering discrete time cn (k) samples with a hn (k) &lt;#. (*) = £ * impulse response, (Oc. (* - ol

[0034] Then modified synthesis (4)

<img file="PL2265041T3_D0003.tif" />

(5) can be calculated in the frequency domain as JM = // (ω) χ (ω), (6) where ^ ω) means Fourier transform f (t) and

<img file="PL2265041T3_D0004.tif" />

(7) [0035] In this case, Hn (ω) = Σkhri (k) exp (-ikω) is a time discrete Fourier transform of the filter applied in the n frequency band for n&gt; 0 i <a name="caption1"></a>///ύ&gt;)=//_,./-ύΐΓ for m &lt;0.

(8) [0036] It should be noted here that the special case Η (ω) = 1 leads to Α (ω) = 1 in (7) due to the special design of the window v (t). Another interesting case here is H &lt;o) = exp (-io), which gives H ^) = exp (-Z ω), so that y (t) = Xt-1).

Proposed solution [0037] In order to obtain a delay of size τ such that yt) = x (t-), the problem is to design filters Ηη (ω) for n&gt; 0 so that

7 / (ω) = exp (-j'ra&gt;), (9) where Η (ω) is given by (7) and (8). The specific solution proposed here is the use of cxp filters (-ijr (rt +1 Z2) r) G_r(Cu)_t for even n exp (-ór (n + l / 2) r) G_c(it) + Jr), for odd n (10) [0038] In this case, &amp; (ω) = &amp; (ω + π) * implies consistency with (8) for all n.

The introduction (10) on the right side of equation (7) gives

H (&amp;) = exp (- / wr) [vol^/r (ii&gt;) G / (v) + K_f&lt;fll + JT) G_r(or + *)] (11) where V () = 2nb (i— (2n + 1/2)) z ό (ω) = exp (i-ra) | y (ω) | Basic calculations show that Vt (ω) is a discrete Fourier transformation

<img file="PL2265041T3_D0005.tif" />

(12)

A very good approximation of the ideal delay can be obtained by solving the linear arrangement <a name="caption2"></a>V_t O) G_f (&lt;u) + (ń&gt; + TT) G_r (ω + Λ-) = 1 (13) least squares method using the filter to be written

FIR. In terms of filter coefficients, equation (13) can

&gt; - /) £, (/) -, 5μ], '' (14)

Where δ [Λ] = 1 for k = 0 and δ [Α] = 0 for k * 0.

[0040] In the case of a discrete-time L-band filter bank with a prototype filter p (k), the obtained delay in the sample units is Lt and the calculation (12) is replaced by ^ (4) = / ^ /, (/) /, (/-7--2.4).

'(15) where T is the integer nearest to Lt. In this case, p (k) is extended by zeros outside the supported range. For a prototype filter with a finite length, only the finite number Vt (k) is different from zero, and (14) is a system of linear equations. The number of unknowns g ^ k) is typically chosen as a small number. In the design of good QMF filter banks, 3-4 branches already provide very good delays. In addition, the dependence of the gT (k) branches of the filter on the delay parameter τ can often be effectively modeled by low-degree polynomials.

Adaptive time grid signaling for stereo parameters [0041] Parametric stereo systems always lead to trade-offs in terms of limited time or frequency resolution to minimize the amount of data transferred. However, it is known from psychoacoustics that some spatial cues may be more important than others, which leads to the possibility of removing less relevant cues. Therefore, the time resolution does not have to be constant. Large gains in throughput can be obtained by enabling time grid synchronization with spatial directions. This can easily be achieved by sending a variable number of parameter sets for each data frame that corresponds to a fixed-size time segment. In order to synchronize parameter sets with the appropriate spatial guidance, additional time grid data describing the location in time for each parameter set must be sent. The resolution of these time indicators can be chosen as quite low in order to keep the total amount of data minimized. The special case in which the time hand for the parameter set coincides with the beginning of the frame could be clearly signaled to avoid sending this time indicator.

[0042] Fig. 4 shows a device according to the invention for performing parameter analysis for time segments having variable and signal-dependent time limits. The device according to the invention comprises means 401 for dividing the input signal into one or several time segments. The time limits that separate the time segments are provided by means 402. Means 402 use a detector specifically designed to acquire spatial cues that are important for deciding where to set time limits. Means 401 output the entire input signal divided into one or several time segments. This output signal is input to means 403 for separate parameter analysis for each time segment. Means 403 output one set of parameters to the analyzed time segment.

[0043] Fig. 5 shows an example of how a time grid generator can behave in the case of a hypothetical input signal. In this example, one set of parameters is used per data frame if no other time limit information is present. Hence, when no other time limit information is present, the natural time limits of the data frame are used. The time limits shown in Fig. 5 are derived from means 402 in Fig. 4. Time segments shown in Fig. 5 are provided by means 401 in Fig. 4.

[0044] The apparatus for encoding the stereo signal to obtain the mono output signal and the stereo parameter set comprises means for calculating the mono signal by combining the channels of the left and right stereo signals in weighted summation. In addition, means 403 generate the first set of stereo parameters using the left channel part and the right channel part, the parts beginning at the first time limit being connected to the means for determining the validity of the first set of stereo parameters for the subsequent left and right channel parts.

[0045] The fixing means are formed together by means 402 and 401 of Fig. 1.

[0046] In particular, the means for determining operate to generate a second time limit and activating the means for generating when it is determined that the first set of stereo parameters is no longer valid, so that a second set of stereo parameters is generated for a portion of the left and right channels starting from the second time limit.

[0047] Not shown in Fig. 4 are means for providing the mono signal, the first stereo parameter set and the first time limit associated with the first stereo parameter set and the second set of stereo parameters and the second time limit associated with the second set of stereo parameters as the parametrically coded stereo signal. Means for determining the validity of a stereo parameter set may include a transient decoder, since it is highly likely that a new stereo parameter must be generated after the transient because the signal has changed significantly in shape. Alternatively, means for determining validity may include a device for analysis by synthesis, which is adapted to decode a mono signal and a set of stereo parameters to obtain a decoded left and decoded right channel in order to compare a decoded left channel and a decoded right channel with a left channel and a right channel and to activate means for generating, when the decoded left channel and the decoded right channel differ from the left channel and right channel by more than the set threshold.

[0048] Data frame 1: The time segment corresponding to the set of 1 parameters begins at the beginning of the data frame 1 because no other time limit information is present in this data frame.

[0049] Data frame 2: Two time limits are present in this data frame. The time segment corresponding to the set of 2 parameters begins at the first time limit in this data frame. The time segment corresponding to the set of 3 parameters begins at the second time limit in this data frame.

[0050] Data frame 3: One time limit is present in this data frame. The time segment corresponding to the set of 4 parameters begins at the time limit in this data frame.

[0051] Data frame 4: One time limit is present in this data frame. This time limit coincides with the initial limit of the data frame 4 and does not need to be signaled because it is supported by default. Hence, the time limit signal can be removed. The time segment corresponding to the set of 5 parameters begins at the beginning of the data frame 4, even without signaling this time limit.

Use of artificial reverberation as a de-correlation method for parametric stereo reconstruction [0052] One critical issue in implementing stereo synthesis in a parametric stereo system is to reduce the coherence between the left and right channels to create a stereo image width. This can be done by adding a filtered version of the original mono signal to the additional signal, where the additional signal and the mono signal are defined by mono = (left + right) / 2, and additional = (left-right) / 2 respectively.

[0053] In order not to change the color too much, the filter should preferably be a full-pass filter. One effective approach is to use a similar full-pass filter used for the artificial reverberation process. Artificial reverberation algorithms usually require high time resolution to provide an impulse response that is sufficiently diffused in time. Basing artificial reverberation algorithms on a complex filter bank, such as the complex QMF bank, has great advantages. The filter bank provides excellent options for enabling reverberation properties to be matched to frequencies in the range of, for example, reverberation correction, decay time, density and color. However, filter bank implementations usually change the time resolution for higher frequency resolutions, which usually makes it difficult to perform a reverberation process with a sufficiently smooth time course. To deal with this, a new way will be to use fractional delay approximation by applying only a phase delay by a factor, for each QMF channel, corresponding to a fixed time delay. This simple method of fractional delay introduces a serious blur of time, which in this case is happily desirable in this case. Time blur contributes to time diffusion, which is highly desirable for reverberation algorithms, and becomes larger as the phase delay approaches pi / 2 or - pi / 2.

[0054] Artificial reverberation processes are for natural reasons processes with an infinite impulse response and offer natural exponential decay. [PCT / SE02 / 01372] indicates that if a reverberation unit is used to generate a stereo signal, the reverberation loss may sometimes be undesirable as soon as the sound ends. However, these unwanted reverberation tails can be easily suppressed or completely removed only by changing the reverberation signal gain. A detector to find the signal ends can be used for this purpose. If the reverberation unit generates artifacts in certain specific signals, e.g. transients, a detector for these signals can also be used to suppress them.

[0055] Fig. 1 shows a device according to the invention for implementing a method of de-correlating signals in a parametric stereo system. The device according to the invention comprises means 101 for providing a plurality of subband signals. The delivery means can be a complex QMF filter bank in which each signal is associated with a subband index.

[0056] The subband signals provided by the means 101 of Fig. 1 are inserted into the means 102 for providing the de-correlated signal 102 and to the means 103 and 106 for modifying the subband signal. The output of the means 102 is input to the means 104 and 105 for modifying the signal, and the output signals of the means 103, 104, 105 and 106 are input to the means 107 and 108 for adding subband signals.

[0057] In the embodiment of the invention described herein, means 103, 104, 105 and 106 for modifying the subband signals adjust the level of the de-correlated signal and the raw signal at the output 101 by multiplying the subband signal by the gain factor, so that each sum of each pair gives a signal with magnitude signal correlated given by control parameters. It should be noted that the gain factors used in measures 103-106 for modification are not limited to positive values. They can also be negative values.

[0058] The output signal from the means for adding subband signals 107 and 108 is input to the time domain receiving means 109 and 110. The output signal from 109 corresponds to the left channel of the reconstructed stereo signal, and the output signal from 110 corresponds to the right channel of the reconstructed stereo signal. In the embodiment described here, the same decorrelator is used for both output channels, and the means for adding the de-correlated signal to the raw signal are separate for the two output channels. The embodiment described here thus ensures that both output signals can be identical and at the same time completely de-correlated, depending on the control data provided by the means for adjusting signal levels and the control data provided to the means for adding signals.

[0059] In Fig. 2 a block diagram of a means for providing a de-correlated signal is shown. The input subband signal is introduced into the means for filtering the subband signal 201. In the embodiment of the present invention described herein, the filtering step is a reverberation unit comprising full-pass filtering. The filter coefficients used are provided by the means for providing filter coefficients 202. The subband index of the currently processed subband signal is entered into 202. In one embodiment of the present invention, different filter coefficients are calculated based on the subband index provided to 202. The filtering step at 201 is based on delayed samples of the input subband signal as well as on delayed samples of intermediate signals in the filtering procedure.

[0060] An essential feature of the present invention is that the means for providing the total delay of the subband sample and the fractional delay of the subband sample are provided by 203. The output signal from 201 is provided to the means for adjusting the level of the subband signal 204 as well as to the means for estimating the properties of the subband signal 205. In a preferred embodiment of the present invention, the estimated properties are transient signal behavior of the subband signal. In this embodiment, the detected transient is signaled to the means for adjusting the subband signal level 204, so that the signal level is reduced during the transient transition. The output signal from 204 is the de-correlated input signal for 104 and 105 in Fig. 1.

[0061] In Fig. 3 a single analysis filter bank and two synthesis filter banks are shown. The 301 analysis filter bank operates on the mono input signal, while the 302 Synthesis filter bank operates on the reconstructed stereo signals.

[0062] Fig. 1 thus, it shows a device according to the invention for generating a de-correlation signal, which is designated 102. As shown in Fig. 1 or 3, the device includes means for providing a plurality of subband signals, wherein the subband signal comprises a sequence of at least two subband samples, the sequence of subband samples represents the bandwidth of the subband signal that is smaller than the bandwidth of the input signal. Each subband signal is input into filtering means 201. Each filter means 201 includes a reverberation filter such that a plurality of reverberated subband signals are obtained, with many of the reverberated subband signals together representing the de-correlated signal. Preferably, as shown in Fig. 2, subband post-processing of the reverberated subband signals may occur, which is performed by block 204, which is controlled by block 205.

[0063] Each reverberation filter is set to a certain delay, preferably a fractional delay, as well as each reverberation filter contains several reverberation factors, which, as shown in Fig. 2, depend on the subband index. This means that it is preferable to use the same delay for each subband, but different sets of filter coefficients for different subbands. This is symbolized by means 203 and 202 in Fig. 2, although it should be mentioned that the delays and filter coefficients are preferably fixed permanently when the de-correlation device is sent, while the delays and filter coefficients can be determined experimentally using listening tests, etc.

[0064] In Fig. 1 a multi-channel decoder is shown and includes a device according to the invention for generating a correlation signal, designated 102 in Fig. 1. The multi-channel decoder shown in Fig. 1 is used to decode a mono signal and associated measure of inter-channel coherence representing coherence between many original channels, while the mono signal is obtained from many original channels. Block 102 in Fig. 1 is a generator for generating the de-correlation signal for a mono signal. Blocks 103, 104, 105, 106 and 107 and 108 constitute a mixer for mixing the mono signal and the decrelated signal according to the first mixing mode to obtain the first decoded output signal and according to the second mixing mode to obtain the second decoded output signal, the mixer works to determine the first mixing mode and the second mixing mode based on the measure of inter-channel coherence sent as additional information for a mono signal.

[0065] The mixer preferably works to mix in the subband domain based on separate measures of inter-channel coherence for different subbands. In this case, the multi-channel decoder further includes means 109 and 110 for converting the first and second decoded output signals from the subband domain in the time domain to obtain the first decoded output signal and the second decoded output signal in the time domain. Hence, means 102 for generating the de-correlation signal according to the invention and the multi-channel decoder according to the invention shown in Fig. 1 operate in the subband domain and, in the last step, convert from the subband domain to the time domain.

[0066] Depending on the actual situation, the device according to the invention can be implemented in hardware or in software or in firmware including hardware and software components. When a program implementation, in whole or in part, takes place, the invention is also a computer program comprising a computer readable code for implementing the methods of the invention when the program is running on a computer.

Dolby International AB, the Netherlands; Proxy:

EP 2 265 041 B1 Z-16974