EP1016319B1 - Process and device for coding a time-discrete stereo signal - Google Patents

Abstract

A method of coding a time-discrete stereo signal, the stereo signal having a first and a second channel, permits scalable stereo coding. At first, a mono signal is formed from the stereo signal, which is then coded, whereupon the coded mono signal is transmitted to a bit stream. Thereafter, the coded mono singal is decoded again, whereupon stereo information is formed on the basis of the coded/decoded mono signal and the first and second channels, with such stereo information being coded and being also written into the bit stream in order to obtain a bit stream comprising a complete coded monolayer as well as a layer with coded stereo information.

Description

The present invention relates to scalable audio encoders and particularly on methods and devices for coding a discrete-time stereo signal.

Scalable audio encoders are encoders that have a modular structure are. So there is an endeavor to already exist Speech encoder to use the signals, e.g. B. with 8 kHz are sampled, process and data rates of, for example Output 4.8 to 8 kilobits per second. This known encoders, such as. B. those known to experts Encoders G. 729, G.723, FS1016, CELP or parametric Models of the MPEG-4 audio VM are mainly used for Encode speech signals and are generally for Coding of higher quality music signals not suitable, as they are usually used for 8 kHz sampled signals are designed, which is why they are only one audio bandwidth of a maximum of 4 kHz. However, they show in generally fast operation and low Computing effort.

For audio coding of music signals, for example Achieving HIFI quality or CD quality is therefore a matter of a scalable encoder a speech encoder with a Audio encoder combined, which signals with a higher sampling rate, such as B. 48 kHz, can encode. Of course it is also possible to use the speech encoder mentioned above to replace another encoder, for example by a music / audio encoder according to the standards MPEG1, MPEG2 or MPEG4.

Such a chain connection of a speech encoder a higher quality audio encoder is commonly used the method of differential coding in the time domain. On Input signal, for example, a sampling rate of 48 kHz, is based on a downsampling filter downsampled the sampling frequency suitable for the speech encoder. Now the down sampled signal coded. The encoded signal can be sent directly to a bit stream formatter be fed to be transmitted. However, it only contains signals with a bandwidth of e.g. B. maximum 4 kHz. The encoded signal is also restored decoded and up-sampled using an upsampling filter. However, the signal now received has due to of the downsampling filter only with useful information a bandwidth of 4 kHz, for example. It should also be noted that that the spectral content of the sampled up coded / decoded signal in the lower band up to 4 kHz not exactly the first 4 kHz band of the one sampled at 48 kHz Corresponds to input signal since encoder in general Introduce coding errors.

As already mentioned, a scalable encoder has a well known speech coder as well an audio encoder that receives signals with higher sampling rates can process. To transmit signal components of the input signal to be able to, whose frequencies are over 4 kHz, a Difference of the input signal with 8 kHz and the coded / decoded, up-sampled output signal of the Speech encoder for each individual time-discrete sample educated. This difference can then be determined using a known Audio encoders can be quantized and encoded as it is for Is known to experts. At this point it should be noted that the difference signal that is in the audio encoder, the signals can code with higher sampling rates, is fed in lower frequency range apart from coding errors of the Speech encoder is much smaller than the original. In the spectral range, which is above the bandwidth of the up-sampled coded / decoded output signal of the speech encoder, the difference signal corresponds to essentially the true input signal, which with z. B. 48 kHz was scanned.

In the first stage, i.e. H. the level of the speech coder, is usually an encoder with a low sampling frequency used because generally a very low bit rate of coded signal is sought. Several are currently working Coders, also called coders, with bit rates of a few Kilobits (two to eight kilobits or more). The same also allow a maximum sampling frequency of 8 kHz, since there is no longer any audio bandwidth at this low level Bit rate is possible, and coding at lower Sampling frequency is cheaper in terms of computing effort. The maximum possible audio bandwidth is 4 kHz and is limited in practice to about 3.5 kHz. Should now in the further stage, d. H. in the stage with the audio encoder, bandwidth improvement must be achieved another stage with a higher sampling frequency. For Adjustment of the sampling frequencies are decimation and Interpolation filter used for downsampling or upsampling.

To date, however, there are only scalable encoders for Mono signals known or implemented. Would be desirable however, a concept for scalable audio encoders which Have joint stereo capabilities. Under "joint stereo" are Stereo coding techniques, such as B. the middle-side coding (M / S coding) or the intensity stereo coding (IS coding) to understand. If just for the left (L) and the right (R) channel of a stereo signal is a separate one scalable mono audio encoder is used, a stereo signal can be encoded, the encoding takes no account of joint stereo techniques, which in the bit-saving coding of stereo signals open up far-reaching savings opportunities can.

The object of the present invention is a Method and device for coding a time-discrete To create stereo signal which use of joint stereo techniques.

This task is accomplished by a method of encoding a discrete-time stereo signal according to claim 1 and by a device for coding a discrete-time stereo signal solved according to claim 14.

The present invention is based on the finding that that a combination of joint stereo techniques with the Principle of scalability can be achieved when out the left and right channels of a stereo signal first a mono signal is formed, which is preferably by summation can happen. The mono signal is generated by means of a first Encoder coded, whereupon the resulting Signal is fed to a bitstream multiplexer. The encoded Mono signal is also decoded again to an encoded / decoded Obtain mono signal that differs from the original Mono signal differs in that it has coding errors having introduced by the first encoder have been. From this encoded / decoded mono signal and the left and right channels of the discrete-time stereo signal stereo information can now be generated, for example Middle / Side (M / S) information or Intensity stereo (IS) information or even under certain The original left channel or the original right channel can be. As follows becomes evident, the coded / decoded can also Mono signal itself or the difference of the original Mono signal from the coded / decoded mono signal as Stereo information used to be used along with the Difference from left and right channel, which also as S signal is called, directly a middle / side coding to surrender. The stereo information can now by means of a second encoder, which is identical to the first encoder or can also be constructed differently from the first encoder, coded and also fed to a bitstream multiplexer, which is a bit stream of the encoded mono signal and the encoded stereo information as well as from later Decoding necessary page information generated.

Forming and encoding the mono signal can take place in the time domain if as the first encoder or Core encoder z. B. a speech encoder is used. Preferably finds the formation and coding of stereo information in the frequency domain, because then on powerful Can be used after the encoder work psychoacoustic model.

However, it is also possible that before further processing the left and right channels are transformed into the frequency domain be, which also leads to coding of the mono signal, a frequency domain encoder can be used which can be done using the psychoacoustic model can encode as distortion-free as possible.

Is used for the first encoder, i.e. H. for the encoder of the mono signal, an encoder is used, which is a lower one Sampling rate has as the time-discrete stereo signal to be encoded, so it has to be from the summation of left and right Channel formed mono signal first to the lower sampling frequency be implemented, which is also known as downsampling becomes. The converted to the lower sampling frequency Mono signal is now encoded and decoded again, whereby the encoded / decoded mono signal is also the lower Sampling frequency. To with the higher sampled left and right channels are related In order to form stereo information, the coded / decoded mono signal back to the sampling frequency of the time-discrete stereo signal, which is also implemented as Upsampling is called. This is done by upsampling obtained coded / decoded mono signal of a frequency domain transformation subjected, which preferably as MDCT (MDCT = modified discrete cosine transformation) can be implemented, the resulting has transformed encoded / decoded mono signal the same time and Frequency resolution like the original discrete-time stereo signal, d. H. the left (L) channel and the right (R) channel.

On the other hand, it becomes the first encoder with the same sampling rate operated, which has the discrete-time stereo signal, so of course, no downsampling and upsampling become.

Fig. 1

einen skalierbaren Stereocodierer mit Monosignalbildung und -codierung im Zeitbereich und Mitte/Seite-Codierung im Frequenzbereich gemäß einem ersten Ausführungsbeispiel der vorliegenden Erfindung;

Fig. 2A

einen skalierbaren Stereocodierer mit Monosignal-bildung und -codierung im Zeitbereich und einer L/R- oder M/S-Codierung im Frequenzbereich gemäß einem zweiten Ausführungsbeispiel;

Fig. 2B

eine detailliertere Darstellung des skalierbaren Stereocodierers von Fig. 2A;

Fig. 3

eine erweiterte Darstellung des skalierbaren Stereocodierers, der in Fig. 2A gezeigt ist, gemäß einem dritten Ausführungsbeispiel der vorliegenden Erfindung; und

Fig. 4

einen skalierbaren Stereocodierer mit Monosignalbildung im Zeitbereich und wahlweiser L/R- oder M/S-Codierung im Frequenzbereich.

Preferred embodiments of the present invention are explained in more detail below with reference to the accompanying drawings. Show it:

Fig. 1

a scalable stereo encoder with mono signal generation and coding in the time domain and center / side coding in the frequency domain according to a first embodiment of the present invention;

Figure 2A

a scalable stereo encoder with mono signal generation and coding in the time domain and L / R or M / S coding in the frequency domain according to a second embodiment;

Figure 2B

a more detailed representation of the scalable stereo encoder of FIG. 2A;

Fig. 3

an expanded representation of the scalable stereo encoder shown in Fig. 2A, according to a third embodiment of the present invention; and

Fig. 4

a scalable stereo encoder with mono signal formation in the time domain and optional L / R or M / S coding in the frequency domain.

Fig. 1 shows a basic block diagram of a scalable Stereo encoder 100 according to a first embodiment of the present invention. The scalable stereo encoder receives a discrete-time stereo signal that a first or left channel L and a second or right channel R includes. First, the stereo signal is preferred by summation by sample by means of a summator 102 a sum signal is formed, which is then by means of multiplier 104 multiplied by a factor of 0.5 is to a mono signal in this embodiment generate that to the middle signal known from the M / S coding is identical. The mono signal at the output of the multiplier 104 is fed into a downsampling filter 106, around the sampling rate to a preferably implement lower sampling rate, which encodes the Mono signal by means of a time domain encoder, which Part of the core codec 108 is to enable. The encoded Mono signal is sent together with corresponding page information written in a bitstream multiplexer 110 which generates a bit stream at its output 112, which encodes a Representation of the discrete-time stereo signal is.

The encoded mono signal is within the core codec 108 decoded again by means of an upsampling filter 114 to be converted back to the first sampling rate so the encoded / decoded mono signal with the left and the right channel for later formation of stereo information can be related.

The discrete-time stereo signal could, for example, by means of a first sampling rate, e.g. B. 48 kHz has been sampled his. The downsampling filter 106 could use this signal the first sampling rate to a second sampling rate of e.g. B. 8 implement kHz. Preferably form the first and the second Sampling rate an integer ratio. The downsampling filter 106 can be implemented as a decimation filter, for example his. The core codec 108 could, for example a speech coder such as e.g. B. G.729, G.723, FS1016, MPEG-4 CELP, MPEG-4 PAR, or a similar encoder. Such encoders operate at data rates of 4.8 kilobits per Second (FS1016) up to data rates of 8 kilobits per second (G.729). However, it is obvious to experts that any other encoder with different data rates or sampling frequencies other than core codec 108 are used can.

If a coder is used as the core codec, which at 8 kHz works, the coded mono signal has a maximum of one Bandwidth of 4 kHz, since the downsampling filter 106 does Mono signal z. B. by decimation to a sampling frequency of 8 kHz. Within the range of 0 - 4 kHz are now the encoded / decoded mono signal and the original one Mono signal at the input of the downsampling filter 106 apart from those introduced by the core codec 108 Coding errors equal. However, it should be noted that the coding error introduced by the core codec 108 is not are always small mistakes, but that they are easily can come in orders of magnitude of the useful signal, if for example a strongly transient signal in the first encoder is encoded. For this reason, as will be discussed later is checked whether differential coding at all makes sense.

The output signal of the upsampling filter 114 now becomes the same like the left and right channels using MDCT filter banks 116 implemented in the frequency domain. The output signals of MDCT filter banks 116, as shown in FIG. 1 is shown, a first frequency-selective switching device (FSS) 118a or a second frequency selective Switching device 118b directly or via a first one Totalizer 120a or a second totalizer 120b indirectly fed.

In particular, the output signal of the MDCT filter bank for the left channel of the first frequency-selective switching device (FSS) 118a, which is also the sum of the transformed left channel and the one with negative Signed transformed coded / decoded Mono signal received. The second frequency-selective switching device 118b receives the next to the transformed R channel Sum of the transformed R channel and the one with negative Signed coded / decoded mono signal.

Check the frequency-selective switching devices 118a, 118b, whether it is more favorable, the transformed original left or right signal or the difference of the left or right signal and the encoded / decoded mono signal to process further. The function of frequency selective Switching device will be shown in more detail later.

The output signal of the first frequency-selective switching device 118a is both a third summer 122a and also a fourth summer 122b with a positive sign fed while the output signal of the second frequency selective Switching device 118b the third summer 122a with a positive sign and the fourth summer 122b with negative sign is supplied. At the exit of the third Summer 122a is now either the sum of the transformed left and right channels or the difference from the Sum of the uncoded left and right channels and the encoded / decoded sum of the left and right channels in front. This signal, which is now in contrast to the encoded Mono signal of the core codec has 108 stereo information, is, for example, by means of an M encoder 124 Consideration of the psychoacoustic model coded and fed to the bitstream multiplexer 110.

However, at the output of the fourth summer 122b Difference of the transformed left and right channel before, this signal also in technology as a side signal is referred to, which is fed into an S-encoder 126 with the S encoder 126 as well as the M encoder 124 code taking into account the psychoacoustic model can. The output signal of the S encoder 126 also becomes fed and included in the bitstream multiplexer also stereo information regarding the time discrete Stereo signal at the input of the scalable stereo encoder 100 according to the first embodiment of the present Invention. It is obvious to experts that a complete bitstream page information needed. Relevant to the invention Side information is especially information of the frequency-selective switching devices 118a and 118b regarding the fact in which frequency band difference signals or transformed L or R signals to the third summer 122a or to the fourth summer 122b have been issued.

In the following, the functions of individual elements, insofar as they have not yet been explained.

The output signal of the core codec 108 points, as it already does was mentioned, e.g. B. a sampling frequency of 8 kHz. This signal, i.e. H. the mono signal, with a lower sampling rate than the original discrete-time stereo signal but now related to the left or right channel brought to form stereo information. To comparable To receive signals, the signal must therefore be included lower sampling rate in a signal with the same sampling rate how the sampling rate of the discrete-time stereo signal is implemented become.

This can be done between the individual Discrete-time samples of the encoded / decoded mono signal a certain number at the output of the core codec 108 of zero values is inserted. The number of zero values is calculated from the ratio of the first and the second Sampling frequency. The ratio of the first (high) to the second (low) sampling frequency is called the upsampling factor designated. However, as is known, the Inserting zeros with very little computation is possible to generate an aliasing disorder that is such affects that the low frequency or zero spectrum of the encoded / decoded mono signal at the output of the core codec 108 is repeated, in total as many times as many Zeros have been inserted. The aliasing signal is now in the frequency range using the MDCT filter bank 116 transformed. By inserting z. B. 5 zeros between Every sample produces a signal, from the start it is known that only every 6th sample of this Signal is different from zero. This fact can Transform this signal into the frequency domain using a filter bank or a modified discrete cosine transform or by means of any frequency transformation be used, for example, for certain Summations that occur with a simple FFT are dispensed with can be. The structure of the signal to be transformed can thus be advantageous How to save computing time when transforming it be used in the frequency domain.

The coded / decoded converted up to the first sampling frequency Mono signal is only in the lower frequency band a correct representation of the original mono signal on Output of the multiplier 104, which is why at the output of the MDCT filter bank 116 only a maximum of one / upsampling factor of the entire spectral lines is used. The insert the zeros in the encoded / decoded mono signal on Output of core codec 108, however, causes the spectral representation of the encoded / decoded mono signal now the same time and frequency resolution as the transformed one has left and right channels.

It is not always convenient to post-process a difference the frequency-selective switching devices 118a and 118b use. The frequency-selective switching devices lead hence a so-called simulcast difference switchover. It is then unfavorable, for example, a difference signal further to process if the difference signal is a higher one Energy than the corresponding other signal at the input of the frequency-selective switching device 118a. There as Core codec 108 any encoder can be used may happen that the encoder by certain the M-encoder 124 or by the S-encoder 126 difficult coding signal components produced. The core codec 108 should preferably phase information of the coded by him Preserve signals, which in the professional world as "waveform coding" or "waveform encoding". The decision, which is the frequency selective switching module 118a or 118b is carried out, preferably frequency-dependent.

"Differential coding" means that only the difference the transformed left or right channel and the transformed encoded / decoded mono signal becomes. If this differential coding is not cheap is because the energy content of the difference signal is greater than the energy content of the transformed left or right Signal is apart from a difference coding and switched to simulcast mode.

Since the difference formation in the frequency domain, i.e. H. selectively by spectral value, it is easily possible a frequency-selective simulcast or differential coding perform. The difference in the spectrum thus allows a simple frequency selective choice of Frequency ranges which are to be differentially coded. In principle, a switch from a differential to a simulcast coding for each spectral value individually occur. However, this would result in too much page information require. Therefore, it is preferred, for example a frequency group comparison of the energies the difference spectral values and the transformed left or right channel. Alternatively can set certain frequency bands from the outset be, e.g. B. 8 bands of 500 kHz each in the example. On There is a compromise in the definition of the frequency bands in the amount of page information to be transmitted, d. H. whether the differential coding is active in a frequency band is or not to weigh against the benefits that come from differential coding as often as possible.

Forming stereo information based on the encoded / decoded Mono signal and the first and second Channel therefore includes a determination of where it's cheaper the transformed left or right channel or one Difference between the same and the encoded / decoded mono signal to process. In each selected frequency band now a frequency-selective comparison of respective energies carried out. If the energy in a certain frequency band the difference signal is the energy of the other Signal multiplied by a predetermined factor k exceeds, it is determined that the output of the frequency-selective switching device 118a the original transformed left signal is. Otherwise it is determined that the difference spectral values are output. The Factor k can range, for example, from about 0.1 to 10. With values of k less than 1, simulcast coding is already used used when the difference signal is lower Has energy than the other signal. At values of k on the other hand, greater than 1, differential coding is still used, even if the energy content of the difference signal already larger than that of the original left or right channel is. As an alternative to the difference formation described can also form stereo information be carried out such that, for. B. a ratio or another link of the encoded / decoded mono signal and the transformed left and right channels is implemented.

2A shows a scalable stereo encoder 200 according to FIG a second embodiment of the present invention. The same elements have the same reference numerals and if they behave the same way, not again described. The scalable stereo encoder 200 differs different from the scalable stereo encoder 100 according to the first embodiment of the present invention in essential in the fact that either a middle / side coding or L / R coding can be performed.

For this purpose, the scalable stereo encoder 200 includes further summing devices 202a, 202b in order to derive from the transformed left and right channel a center signal M or to generate a side signal S. That transformed encoded / decoded mono signal is referred to here as M '. The signal M and the signal M 'is also additional frequency-selective switching device 204 fed in, which generates a signal M ″, the frequency-selective Switching device 204 also a summer 206 is connected upstream, as is the case with all other frequency-selective Switching devices is the case. The scalable Stereo encoder 200 also includes a block joint stereo decision 208, which 4 input signals L ', M ", S and R 'receives. The block joint stereo decision 208 decides in a known way, whether from a stereo encoder 210 an L / R, an M / S or an intensity coding is to be carried out.

The function of the scalable stereo encoder 200 is shown below. First, a mono signal is formed from the discrete-time stereo signal, this formation taking place in the time domain and equatingly as follows: M T = (L + R) 0.5

The index T is intended to indicate that this is a middle signal in the time domain. The core encoder 108 now operates as was shown in connection with FIG. 1. In addition, as in FIG. 1, an MDCT is also carried out on the L and R signals. The M / S signal in the frequency range is now calculated using the summers 202a and 202b and the downstream multipliers, which is expressed as follows in equations: M = (L + R) x 0.5 and S = (L - R) 0.5

The frequency-selective switching device now serves as it has already been mentioned for calculating M ''. Damn either equal to M - M 'or M itself, as already shown has been. The frequency selective switching device 118 computes the signal L ', which is either equal to 0.5 (L - M') or equal to 0.5 · L. The same applies to the signal R ', which is either equal to R · 0.5 or equal to (R - M') · 0.5 is. The switching devices 118a, 118b and 204 operate frequency selective. In the block joint stereo decision 208 a decision is now made in the usual way whether a coding of the signals L 'and R' or M "or S has to take place. This function is known in the art and is therefore not explained in more detail.

FIG. 2B shows a scalable stereo encoder which differs in some points from the scalable stereo encoder 200 according to the second exemplary embodiment of the invention. As the only multiplier, the same comprises the two multipliers 214a and 214b, which are arranged after the frequency-selective switching device 204 and after the frequency-selective switching device 118b. 2B also includes a somewhat more detailed illustration of the frequency selective switching devices. The switch state of the frequency selective switch 118a, which is referred to as S _1LR , will always be complementary to the switch state of the frequency selective switch 118b, which is referred to as S ' _1LR . The same applies to two additional switches S ₂ and S ₂ ', which may be present in the joint stereo decision block 208 to form internal signals L "and R".

Moving the multiplications behind the frequency-selective switching devices leads to a simpler and clearer representation of the stereo encoder. The multiplications per se are therefore no longer absolutely necessary, but the same could also be carried out in the decoder. In order to reduce the side information to be transmitted, it is also possible to transmit only a few switch states instead of the transmission of all switch states. If the switch S ₂ indicates the state a that L / R coding is used, it is sufficient to transmit only the state of the switches S ₁ , S ' ₁ , the transmission of the state of the switch S' ₁ can be omitted , since this will be complementary to the state of the switch S ₁ . If S ₂ assumes a different state, ie state b, as shown in the drawing, it is sufficient to transmit the state S _{1M of} the frequency-selective switching device 204, which indicates whether differential or simulcast coding of the signal M is carried out. If the switch S _{2 is} in a position c, side information is transmitted that there is an intensity stereo coding, in which case the position of the switch S _{1M is also} transmitted, while here the positions of S _1LR and S ' _1LR are irrelevant.

3 includes another embodiment 300 of a scalable stereo encoder according to the present invention. The embodiment shown in Fig. 3 differs differs from the embodiment shown in FIG. 2 essentially in that the mono signal in two Levels is encoded. The first stage is through the core codec 108 formed during the second stage by one Encoder / decoder 302 is formed, which in the preferred embodiment works in the frequency domain and implemented as a psychoacoustic frequency domain encoder can be. It receives the input signal M " Output signal of the frequency-selective switching device 204, it is also checked here whether differential coding or simulcast coding makes sense or not. The output signal of encoder / decoder 302 becomes a summer 304 supplied, the output signal M '' 'the difference of the Signal M and the output signal of the encoder / decoder 302 corresponds. This signal M '' 'is just like that Signals L ', S and R' of a joint stereo decision (not shown) and then a stereo encoder (also not shown) supplied. The core codec 108 also includes that encoder / decoder 302 an output to the bit stream multiplexer, to transmit encoded data to the same. The Outputs of the frequency-selective switching devices to the Bitstream multiplexers are intended to illustrate that page information of frequency-selective switching devices regarding the use of differential and simulcast coding in a frequency band also the bit stream multiplexer need to be fed to a trouble-free To enable decoding. The bit stream includes in 3 additionally shows scalable stereo encoder 300 to the first layer or the first layer that is created by the encoded mono signal of the core codec 108 is formed, a second layer by the coded signal M '' at the bitstream multiplexer output of encoder / decoder 302 is formed, the encoder 300 shown in FIG. 3 an encoding of the mono signal at full sampling rate can enable.

In contrast to the exemplary embodiments described so far 4 illustrates a scalable audio encoder 400 represents a mono signal formation only in the frequency domain performs. For this purpose, the signals L and R by means of MDCT filter banks 116 transformed into the frequency domain, after which an M / S matrix by means of summers 202a and 202b and the subsequent multiplier by a factor of 0.5 is carried out. The multiplier is thus at the output on the one hand a center signal M and on the other hand a side signal S on. The middle signal used as a mono signal can be, is by means of a first encoder / decoder 402 encoded and decoded again, the encoded mono signal M is written in the bitstream, as it already is has been mentioned several times. Downstream of encoder / decoder 402 is a summing device 404 which the Difference between the encoded / decoded mono signal and forms the original mono signal M, this difference is designated as M '. The signals L ', M', S and R ' can again use a joint stereo decision facility are supplied, which, however, not shown in Fig. 4 is.

The encoder 400 presented in FIG. 4 thus operates completely in the frequency domain, the encoder / decoder 402 preferably as a frequency domain encoder full sampling rate is executed. The stereo encoder (not shown) after the IS decision level (also in FIG. 4 (not shown) is preferably also used as a frequency domain encoder executed at full sampling rate. The in Fig. 4 scalable stereo encoder thus represents a Generalization of the term "scalability" because the Bitstream here no layers or "layers" with different Audio bandwidths but (like the other exemplary embodiments) comprises a monolayer and a stereo layer, which encodes separately from one another by an encoder can be. An older monodecoder that doesn't is equipped for stereo operation, for example the bit stream of the encoders according to the invention decode to generate at least one mono audio signal. The scalable stereo encoders according to the invention are thus backwards compatible with existing monodecoders.

Claims (14)

1. A method of coding a time-discrete stereo signal, with the stereo signal having a first and a second channel (L, R), said method comprising the following steps:

   (a) forming a mono signal (M) from the stereo signal;

   (b) coding the mono signal and transmitting the coded mono signal to a bit stream;

   (c) decoding the coded mono signal;

   (d) forming stereo information on the basis of the coded/decoded mono signal (M') and the first and second channels (L, R); and

   (e) coding the stereo information and transmitting the same to the bit stream.
2. The method of claim 1, in which the time-discrete stereo signal has a first sampling rate, wherein step (a) comprises the following partial steps:

   (a21) summing the left-hand and right-hand channels (L, R) by sampling values in order to obtain a sum signal; and

   (a22) converting the sum signal to a second sampling rate lower than the first sampling rate in order to obtain the mono signal; and

   step (c) comprises the following partial steps:

   (c21) decoding the coded mono signal having the second sampling rate; and

   (c22) converting the coded/decoded mono signal to the first sampling rate.
3. The method of any of the preceding claims, further comprising the following step:
   transforming the left-hand and right-hand channels and the coded/decoded mono signal to the frequency domain, with the transformed signals all having substantially the same time and frequency resolution.
4. The method of claim 3, wherein step (d) comprises the following partial steps:

   (d41) frequency-selectively comparing of the transformed left-hand channel to the difference of the transformed left-hand channel and the transformed coded/decoded mono signal, and selecting the signal having the lower entropy in terms of hearing or the lower energy or adapted to be coded with a lower bit number;

   (d42) frequency-selectively comparing of the transformed right-hand channel to the difference of the transformed right-hand channel and the transformed coded/decoded mono signal, and selecting the signal having the lower entropy in terms of hearing or the lower energy or adapted to be coded with a lower bit number;

   (d43) summing the signals selected in steps (d41) and (d42) in order to obtain a mid signal (M) as first stereo information; and

   (d44) subtracting the signal selected in step (d42) from the signal selected in step (d41) in order to obtain a side signal (S) as second stereo information.
5. The method of any of claims 1 to 3, wherein step (d) comprises the following partial steps:

   (d51) summing the transformed left-hand channel (L) and the transformed right-hand channel (R) in order to obtain a mid signal (M); and

   (d52) subtracting the transformed right-hand channel (R) from the transformed left-hand channel (L) in order to obtain a side signal (S).
6. The method of claim 5, wherein step (d) further comprises the following partial steps:

   (d61) frequency-selectively comparing of the transformed coded/decoded mono signal (M') to the difference of the mid signal (M) and the coded/decoded mono signal (M'), and selecting the signal with lower energy;

   (d62) frequency-selectively comparing of the left-hand channel to the difference of the left-hand channel (L) and the transformed coded/decoded mono signal (M'); and

   (d63) frequency-selectively comparing of the right-hand channel to the difference of the right-hand channel (R) and the transformed coded/decoded mono-signal (M').
7. The method of claim 6, wherein step (d) further comprises the following partial step:
   (d71) deciding whether the results of steps (d61) and (d52) or the results of steps (d62) and (d63), respectively, are used as first and second stereo information.
8. The method of claim 7, wherein step (d) prior to step (d71) further comprises the following partial step:
   (d81) halving the results of steps (d61) and (d52).
9. The method of claim 7 or 8, wherein step (d) further comprises the following partial step:
   (d91) if in steps (d71) the results of steps (d62) and (d63) are used as first and second stereo information, transmitting side information indicating either the result of step (d62) or of step (d63), otherwise transmitting side information indicating the result of step (d61).
10. The method of any of claims 1 to 5, wherein step (d) further comprises the following partial steps:
    (d101) frequency-selectively comparing of the mid signal (M) to the difference of the mid signal (M) and the transformed coded/decoded mono signal (M'), and selecting the signal with lower energy as additional mono signal;
    wherein step (b) further comprises the following step:

    (b101) coding the additional mono signal (M") and transmitting the coded additional mono signal to the bit stream; and

    (b102) decoding the coded additional mono signal.
11. The method of claim 10, wherein step (d) comprises the following partial steps:

    (di11) subtracting the coded/decoded additional mono signal (M'') from the mid signal (M);

    (d112) frequency-selectively comparing of the transformed left-hand channel (L) to the difference of the left-hand channel and the result of step (d111), and selecting the signal with lower energy;

    (d113) frequency-selectively comparing of the transformed left-hand channel (L) to the difference of the right-hand channel and the result of step (d111), and selecting the signal with the lower energy; and

    (d114) deciding whether the results of steps (d111) (M''') and (d52) (S) or the results of steps (d112) (L') and (d113) (R'), respectively, are used as first and second stereo information.
12. The method of claim 1, wherein prior to step (a) the left-hand and right-hand channels are transformed ot the frequency domain, with step (a) comprising the following partial step:
    (a121) summing the transformed left-hand and right-hand channels by spectral values in order to obtain the mono signal (M).
13. The method of claim 12, wherein step (d) comprises the following partial steps:

    (d131) subtracting the coded/decoded mono signal from mono signal (M);

    (d132) subtracting the transformed right-hand channel (R) from the transformed left-hand channel (L) in order to obtain a transformed side signal (S);

    (d133) comparing, by spectral values, the transformed left-hand signal (L) to the difference of the transformed left-hand signal (L) and the result of step (d131), and seleting the signal with lower energy;

    (d134) comparing, by spectral values, the transformed right-hand signal (R) to the difference of the transformed right-hand signal and the result of step (d131), and selecting the signal with lower energy; and

    (d135) deciding whether the results of steps (d133) (L') (M') and (d134) (R') or the results of steps (d131) (M') and (d132) (S) are used as first and second stereo information.
14. An apparatus (100; 200; 300; 400) for coding a time-discrete stereo signal, the stereo signal having a first and a second channel (R, L), said apparatus comprising:

    (a) means (102, 104; 202a) for forming a mono signal from the stereo signal;

    (b) means (108; 402) for coding the mono signal and transmitting the coded mono signal to a bit stream;

    (c) means (108; 402) for decoding the coded mono signal;

    (d). means (116, 118a, 118b, 120a, 120b, 122a, 122b; 202a, 202b, 204, 208; 214a, 214b; 302, 304; 402, 404) for forming stereo information on the basis of the coded/decoded mono signal and the first and second channels; and

    (e) means (124, 126; 210) for coding the stereo information and for transmitting the same to the bit stream.

Images