DE3734084C2 - - Google Patents

Description

The invention relates to a method according to the preamble of claim 1. Such The procedure is from IEEE Transactions on Consumer Electronics, Vol. CE-21, 1975, No. 1, pp. 95-103.

To play stereo signals after transmission, it is known to reproduce a decoding matrix to use, which encoded the incoming stereo signals M = L + R and S = L-R decoded to L = M + S and R = M-S. Other encodings too are known (DE-PS 28 27 159). So z. B. the TV audio signals in the coding M = L + R and R transfer. These encodings are necessary in order to to achieve so-called backward compatibility, d. H., that a non-stereo receiver the mono signal receives, and the stereo-compatible receiver with the help a stereophonic supplementary signal by matrixing which can produce stereo L / R speaker signals.

To use two monophones instead of one stereo To be able to carry out transfers, e.g. B. bilingual Television sound signals, it is also known (Funkschau 1982, H. 20, p. 65 and DE-PS 29 02 933), the effect of the decoding matrix in Dependence on a switching signal that is also transmitted change in such a way that the listener on the playback side either the first or the second incoming monophonic Sound signal for playback via the left and right Speaker can choose.  

There are also stereophonic recording / playback processes known, which are based on more than two playback channels (IEEE Transactions on Consumer Electronics, Vol. CE-21, 1975, No. 1, pp. 95 to 103). Which includes

• - The "Triphonic process", which between the front left and right speakers provides additional center speaker in front of the listener;
• - The "Quadrofonie procedure", which besides the front stereo speakers an additional Stereo speaker pair behind the handset required;
• - The "Dolby Surround process", which like that "Triphonic method" an additional center speaker as well as additional speakers on the side of the handset and behind the handset.

In order for these procedures to be compatible with the usual Achieving two-channel stereo becomes special Matrix networks are used, which depend on on the playback side generated characteristic signals the three or four-channel stereo audio signals in a two-channel Transfer stereo signal, and which one if necessary playback-side recovery of the additional tone signals allow (3-2-3 matrix or 4-2-4 matrix). Such However, matrix networks basically cause Quality impairments, namely:

• - The audio signals cannot be completely decoupled and contain crosstalk parts;
• - arise through the use of all-pass networks Phase distortions;
• - by using dynamic matrix control circuits arise to improve the channel separation Distortions caused by control processes.

The impairments mentioned are individual or combined - depending on the matrix type - available and generated playback-side disturbing changes in the sound image.

To avoid these matrix networks it is necessary the number of transmission channels the number of playback channels adapt. As for future digital sound transmission systems effective data reduction procedures can be used (DE-PS 34 40 613), the number of Transmission channels with the same data rate and without perceptible quality impairments, for example be increased from two to ten so that recording / Playback method that has more than two playback channels underlying, without application of Have matrix networks used. Such system concepts are known, for example for the sound representation on high-definition television (DE-PS 34 42 388). With these known system concepts, however, is done for reasons of the necessary compatibility to the usual two-channel stereo playback of a down to a specific recording / reproduction process, such  that a certain transmission channel one is assigned to a specific playback channel. The A specific audio signal is identified with the help of this fixed channel assignment, so that for the In the case of a reduced number of playback channels, for example, playback over two speakers in the usual stereophonic arrangement, the combination several sound signals according to the fixed assignment of the transmission channels. Especially for the Case of sound on high-definition television is this fixed assignment of transmission and playback channels disadvantageous because the selected recording / Reproduction process of a variety of requirements must do justice. These are:

1. Backward compatibility

Because the high-definition television (HDTV) than future Universal standard for film and television productions is provided must be from the HDTV source material highest quality finely graduated, lower quality standards be derivable so that with a reduced number the channels are not associated with unnecessary quality losses are. For example, sound events should be behind in the absence of the rear speakers in adequate form (e.g. through simulation techniques) the front speakers are shown.  

2. System compatibility

The common sound systems for cinema films should can be used without adaptation problems, d. H. the Most of the existing cinema productions should transmitted and reproduced as "true to the original" as possible can be. Future sound systems should can be easily introduced.

3. Stability of the direction map

Within a certain listening zone, such as the Width of the picture should be sufficient Guaranteed congruence between the visual and the auditory event be.

4. Efficient room mapping

The representation of the acoustic environment of the picture (e.g. sound sources outside the image) and the acoustic atmosphere should be in one opposite usual two-channel stereo expanded image range respectively.

5. Flexibility on the receiving side

On the production side, there should be a degree of freedom Use of the audio transmission channels to be in place suitable recording / playback methods depending on the program type to be able to choose which, for example, for Music broadcasts a high quality spatial image, a stable picture-sound congruence for feature films  and effects, or for reporting primarily allow many foreign languages. A "ideal" universal recording / playback method is not feasible.

6. Flexibility on the playback side

The playback device should meet the needs the listener can be adjusted, in particular regarding effort and intention. For example is more important to a big one in the HDTV small cinema Listening zone placed than in the living room. Elsewhere u. U. less importance was attached to the exact picture-sound congruence than on a natural mapping of the acoustic The atmosphere.

It is the object of the invention, on the one hand, flexibility to achieve while maintaining compatibility to ensure that playback side an individually adaptable sound representation also for different types of stereophonic recording methods taking into account the existing ones Playback facility is possible.

This object is achieved by the characterizing Features of the claim solved.

Advantageous refinements and developments of the inventive method result from the Subclaims.

The invention is illustrated by the drawings explained. It shows

FIG. 1 shows schematic representations of various examples of spatial speaker arrays on the reproduction side;

Fig. 2 is a schematic representation of a first embodiment for carrying out the inventive method;

Fig. 3 is a schematic representation of a second embodiment for performing the method according to the invention.

The method according to the invention takes into account the knowledge that a flexible, production-controlled assignment of transmission and playback channels is necessary if, with a limited number of transmission channels, the sound representations are to meet a variety of different requirements. If the identification of a specific sound signal is carried out with the aid of a co-transmitted identification signal, then depending on this sound signal can be assigned to a specific playback channel or a specific loudspeaker. It is assumed that a limited number of different spatial loudspeaker arrangements is standardized, for example for the playback side the arrangements A to H according to FIG. 1. By choosing one of the arrangements, the listener can not only the technical effort (number of playback channels), but also determine the performance characteristics of the display device. For example, four loudspeakers can be set up according to the arrangements D or E according to FIG. 1, depending on whether emphasis is placed on high directional stability or on natural reproduction of the acoustic atmosphere.

If the original loudspeaker arrangement used on the production side is not available on the reproduction side, the sound signals can be adapted to the existing loudspeaker arrangement on the reproduction side in accordance with the characteristic signal transmitted as well. If, for example, according to FIG. 2, the loudspeaker arrangement H according to FIG. 1 is used on the production side and only the loudspeaker arrangement C is available on the reproduction side, then the sound signal adaptation takes place in a signal processing circuit 10, hereinafter referred to as “sound signal combiner”, in accordance with a signal generated on the transmission side, for example by the director 20 , transmitted with the identification signal K _s and in accordance with a reproduction side, generated by the listener identification signal K _w .

The operations of the audio signal combiner 10 required for the audio signal adaptation are comparable to the basic operations of a mixer; they are

in which
W _x the sound signal for the loudspeaker x
S _i the transmitted audio signal i
a _{xi is} a coefficient for S _i when determining W _x with

0 / a _xi / 1

mean. The combination _rules can be fully described by specifying all the coefficients a _xi . The suitable coefficients a _xi for the audio signal adaptation "H to C" required in FIG. 2 by the audio signal combiner 10 are, for example:

a _xi for "H to C"

The resulting operations are:

W₁ = 0.71 S₄ + S₅ + S₆
W₂ = 0.71 S₃ + 0.71 S₄
W₃ = S₁ + S₂ + 0.71 S₃
W₄ = W₅ = W₆ = 0

The second example is the coefficient set for the Case indicated that the reproduction side as well as the production side the same speaker arrangement H used becomes:

a _xi for "H to H"

The third example gives the coefficient set for the case that the loudspeaker arrangement on the production side H and playback side the usual stereophones Speaker arrangement B (or headphone reproduction) are used (W₁ = left channel, W₂ = right Channel):

a _xi for "H to B"

The particular advantage of the signal-controlled tone signal combiner 10 arises not only from the fact that optimum downward compatibility and, at the same time, optimum flexibility are guaranteed on the production side, but also that the existing transmission channels can be used optimally when a multilingual program is to be transmitted. For example, the flexible use of a total of six transmission channels enables the provision of six alternative monoprograms (arrangement A according to FIG. 1, for example alternative languages), or of three alternative stereo programs (arrangement B), or of two alternative "triphonic" - without problems. Programs (arrangement C) in that the corresponding coefficients are alternatively rated by the listener as "zero" or "one". The coefficient set for the case of three alternative stereo programs is e.g. B .:

The coefficients h₁, h₂, h₃ can be zero from the listener or be set to one, the rule "either h₁ or h₂ or h₃ equals one" advantageous is.

The available transmission channels can be used even more effectively with the aid of the audio signal combiner 10 if alternative languages are to be provided by the audio signal components “dialog” (ie each of the languages) on the one hand and “effect” (ie ambient noise) and (underlying) “music "on the other hand, are transmitted separately, so that the final mixing of these transmitted audio signal components can take place on the reproduction side with the aid of the audio signal combiner 10 . For example, four out of six existing transmission channels for alternative languages and the other two channels for stereophonic reproduction of the acoustic atmosphere can be used. B. seems attractive for a realistic sports report. The appropriate set of coefficients for this application example would have the following form, if the loudspeaker arrangement B is used on the reproduction side:

The coefficients h₁ to h₄ represent the alternative Choice of language. Is an additional on the playback side Center speaker available, e.g. B. in the arrangement C, F or G, so the selected dialog signal be emitted via this. For that he would have Coefficient set the form:

(The signals W₁ and W₂ are for the left or right, the signal W₃ for the center speaker arrangement C).  

For operational reasons, it is expedient not only to standardize the loudspeaker arrangements on the reproduction side, but also the loudspeaker arrangements that can be used on the production side. With reference to the examples for the playback page shown in FIG. 1, the restriction appears for the production e.g. B. on the arrangements B, C, E, G, H useful. It is also advisable to limit and standardize the various options for transmitting different programs or alternative languages. Alternative definitions of the signals to be transmitted S₁. . . S₆, which could be of particular interest in the production or broadcast studio, are summarized in Table 1.

In Table 1, the definitions (examples) mean:

B 1 / -30 ° signal for loudspeaker -30 ° of arrangement B ( Fig. 1), alternative program 1
Z 1 Alternative signal, alternative language 1
Z 1 / -30 ° Alternative signal for loudspeaker -30 ° of arrangement B, alternative language 1
C / 0 ° signal for loudspeaker O of arrangement C, no alternative program available
C 2 / -30 ° signal for loudspeaker -30 ° of arrangement C, alternative program 2

Table 1

The alternatives 1 to 9 offer a sufficiently flexible use of the transmission capacity. On the other hand, all the required coefficient sets can be saved by specifying, for example, according to Table 1, so that only one of the alternatives 1 to 9 can be selected on the production or transmission side. Since a suitable sound signal adaptation should be possible for each of the eight standardized reproduction-side loudspeaker arrangements according to FIG. 1, a maximum of 72 sets of coefficients would have to be defined and stored. Depending on the identification signal K _s ( FIG. 2), which identifies one of the alternatives 1 to 9 (Table 1), and depending on the identification signal K _w ( FIG. 2), which identifies one of the loudspeaker arrangements A to H ( FIG. 1) and, if appropriate characterizes the alternative program selected by the listener or the selected alternative language, the sound signal combiner 10 calls up the appropriate coefficient set. Compatibility is thus achieved for all standardized loudspeaker arrangements. At the same time, flexibility is achieved to a high degree without operational problems. This is of particular interest for HDTV sound since the source material has to meet a variety of requirements.

Various embodiments of the method according to the invention are conceivable. For example, the storage of all alternative coefficient sets in the audio signal combiner 10 can be avoided in such a way that any coefficients a _xi are generated on the transmission side and transmitted with the identification signal K _s . In this case, the identification signal K _s must contain the appropriate coefficient sets for each of the standardized, reproduction-side loudspeaker arrangements, so that the appropriate coefficient set can take effect in the audio signal combiner 10 in accordance with the identification signal K _w . In the control room 20 , this embodiment also does not cause any operational problems if the coefficient sets are stored there and, as described above, the alternative definitions for S 1 to S 5 are meaningfully restricted, for example according to Table 1. This exemplary embodiment has the advantage that the mode of operation of the Tone signal combiner 10 is not limited in principle, and that future recording / playback methods can therefore be introduced without compatibility problems by changing or supplementing the alternative coefficient sets stored on the transmission side. However, a higher flow of information is necessary for the control of the audio signal combiner 10 . However, since the coefficients a _xi can be quantized relatively roughly, with a resolution of 64 levels, only about 2 kbit / s are required for the transmission of eight sets of coefficients including their identification codes.

As a further exemplary embodiment, operations which deviate therefrom can be provided in the coupling network of the audio signal combiner 10 described by equation (1), in such a way that the combination rules are not described solely by the coefficients a _xi . In addition to the intended multiplications, delays and / or changes in the frequency response (equalization) are particularly useful in order to optimally design certain sound signal adaptations. By delays and equalizations can be achieved, for example, that sound signals, which are provided to represent the acoustic atmosphere in the arrangement C for the speakers C / -30 ° and C / + 30 °, when using a more complex arrangement G additionally via the speakers G. / -120 °, G / + 120 °, G / 180 ° can be emitted. The acoustic atmosphere appears more spatial, since localization of the rear speakers in arrangement G is avoided as a result of the delays. Such properties of the sound signal combiner 10 can be implemented with a correspondingly greater outlay. Conversely, the sound signal combiner 10 can also be carried out very simply, for example in that only the coefficients a _{xi are} provided and some of them have a fixed value of one, and also in that the coefficients a _xi can only take on a very limited number of values, e.g. . B. the values 0; 0.1; 0.32; 0.52; 0.71; 0.85; 1; -0.71.

As a special application of the method according to the invention, it is expedient to use it in a broadcasting studio where, for example, it is necessary to adapt the HDTV source material with six audio channels to a satellite transmission link for which there are only four audio transmission channels and, accordingly, arrangements of a maximum of four loudspeakers. In this case, the right half of the arrangement according to FIG. 2 is formed by the broadcasting studio, where a signal adaptation from six channels to four channels and a corresponding reduction and adaptation of the coefficient sets takes place.

An advantageous embodiment of the method according to the invention is shown in Fig. 3. Before transmission, the digital audio signals S₁. . . S₆ in coders 30 subjected to data reduction, for example according to DE-PS 34 40 613. After the transmission, the coded audio signals S ₁ ,. . . S₂ 'to the sound signal combiner 10 , which performs the sound signal adaptation in accordance with the identification signals K _s and K _w . Only the output signals W₁ '. . . W₂ 'of the sound signal combiner are decoded. This has the advantage that only as many decoders as speakers are necessary on the playback side. The technical outlay for playback arrangements with a few channels is correspondingly less than when the decoders are used before the audio signal combiner. By using known data reduction techniques, the information flow of a six-channel audio signal can be reduced from 4.2 Mbit / s to about 0.5 Mbit / s. In this case too, the flow of information required for the transmission of the identification signal K _s is negligible.

Claims (7)

1. Method for reproducing multi-channel audio signals, in particular television audio signals for high-definition television, in which identification signals are also transmitted, which information for

• a) the type of the stereophonic recording method on which the sound signals are based, and / or
• b) the combination rules for the audio signals resulting from the type of recording method in different types of display devices

contain, which, together with characteristics generated on the reproduction side, control a signal processing circuit with regard to the type of the respective reproduction device, characterized in that the matrixing on the transmission side via a multiplicity of transmission channels reproduces sound signals on the reproduction side in accordance with the identification signals (K _s ) also transmitted and the identification signals generated on the reproduction side ( K _w ) can be adapted to the number of available playback channels or the existing playback arrangements or the desired listening conditions by mixing (combining) according to the following basic operation in the signal processing circuit ( 10 ): in which
W _x the sound signal for the loudspeaker x
S _i the transmitted audio signal i
a _{xi is} a coefficient for S _i when determining W _{x with} 0 / a _xi / 1. 2. The method according to claim 1, characterized in that the appropriate coefficient sets of a _xi for all standardized playback arrangements are stored in the signal processing circuit. 3. The method according to claim 1, characterized in that the appropriate sets of coefficients of a _xi for all standardized display devices are generated on the transmission side and transmitted with the identification signals (K _s ). 4. The method according to any one of claims 1 to 3, characterized in that the playback-side generated identification signals (K _w ) further contain information about signal delays and / or equalizations. 5. The method according to any one of claims 1 to 4, characterized in that the co-transmitted identification signals (K _s ) further contain information about the program type of the sound signals. 6. The method according to any one of claims 1 to 5, characterized characterized in that the sound signals are digitized and are reduced in bit rate.