HK1030129B

HK1030129B - Method, apparatus and system for processing an audio signal

Info

Publication number: HK1030129B
Application number: HK01100663.5A
Authority: HK
Inventors: J‧理查德‧艾尔沃德
Original assignee: 伯斯有限公司
Priority date: 1998-12-24
Filing date: 2001-01-30
Publication date: 2005-12-02

Description

Audio signal processing method, device and system

The present invention relates to processing audio signals, and in particular to processing one or more audio input signals to provide a plurality of audio signals.

It is an important object of the present invention to provide an audio signal processing system for providing a plurality of audio channel output signals from one or more input signals.

According to the present invention, a method for processing a single channel audio signal to provide a plurality of audio channel signals comprises splitting the single channel audio signal into a first split signal and a second split signal characterized by spectral general characteristics of speech; generating a first channel signal from the first split signal; and modifying the second split signal to produce a remaining channel signal of the plurality of channel signals.

In another aspect of the present invention, an audio signal processing apparatus for processing a single channel audio signal to provide a plurality of audio channel signals includes a speech separator for separating the audio signal into a first separated signal and a second separated signal characterized by spectral general characteristics of speech; and circuitry coupled to the voice splitter for generating a first subset of the plurality of audio channel signals from the second split signal.

In another aspect of the invention, an audio signal processing system includes an input for a single input channel signal; a center channel output for a center channel output signal C; a plurality of outputs for a respective plurality of output path signals; a voice splitter, coupled to the input and the center channel output, for splitting a single channel input signal into a voice audio signal and a non-voice audio signal; and a circuit coupled to the voice splitter to a plurality of outputs for providing a corresponding plurality of audio channel signals at the outputs in response to the non-voice audio signal.

In another aspect of the invention, a method for processing a signaling channel audio signal to provide two decodable audio channel signals that are decodable into five audio signals comprises: separating the single channel audio signal into a first separated signal and a second separated signal characterized by spectral general characteristics of speech; processing the first split signal to provide a center channel signal C; modifying the second split signal to provide a left channel signal L, a right channel signal R, a left surround channel signal Ls and a right surround channel signal Rs; combining the center channel signal, the left surround and right surround channel signals and the sum of the left channel signals to produce a first audio channel signal of two decodable audio channel signals; and combining the center channel signal, the sum of the left surround and right surround channel signals and the right channel signal to produce a second audio channel signal of the two decodable audio channel signals.

In another aspect of the present invention, a method for processing a single channel audio signal to provide three decodable audio channel signals that are subsequently decodable into five audio signals comprises: dividing the single channel audio signal into a first split signal and a second split signal characterized by spectral general characteristics of speech; processing the first split signal to provide a center channel signal, the center channel signal comprising a first decodable audio signal; modifying the second split signal to provide a left channel signal, a right channel signal, a left surround channel signal and a right surround channel signal; combining the left surround and right surround channel signals and the left channel signal to produce a second audio channel signal of the three decodable audio channel signals; and combining the sum of the right surround and right surround channel signals and the right channel signal to produce a third audio channel signal of the three decodable audio channel signals.

In another aspect of the invention, a method for processing two input channel audio signals to provide two or more output audio channel signals comprises: dividing each of two input audio channels into a first split signal and a second split signal characterized by spectral general characteristics of speech; combining the first split signal of the first input audio channel signal and the first split signal of the second input audio channel signal to form a first one of the two or more output audio channel signals; the second split signal of the first input signal is transmitted as a second output audio channel signal of the two or more output audio channel signals, and the second split signal of the second input signal is transmitted as a third output channel signal of the two or more output channel signals.

In yet another aspect of the present invention, an audio signal processing apparatus for processing two input channel audio signals to provide two or more output audio channel signals comprises: a first voice separator for separating a first audio channel signal of the two input audio channel signals into a first separated signal characterized by spectral characteristics of voice to provide a first output audio channel signal of the two or more output audio channel signals; a second voice separator for separating a second audio channel signal of the two audio channel signals into a first separated signal characterized by a spectral characteristic of voice and a second output audio channel signal of the two or more output audio channel signals; and a combiner for combining the first and second split signals to form a third one of the two or more output audio channel signals.

Other features, objects, and advantages will become apparent from the following detailed description, which proceeds with reference to the accompanying drawings. Wherein:

FIG. 1 is a block diagram of a single channel audio signal processing system according to the present invention;

FIGS. 2a and 2b are circuit diagrams of circuits implementing the voice splitter and multi-channel emulator of FIG. 1;

3a-3c are block diagrams of alternative embodiments of the post-simulation processing system of FIG. 1; and

fig. 4 is a circuit diagram of a circuit implementing the principles of the present invention in a two input channel system.

Referring now to the drawings, and more particularly to FIG. 1, a single channel audio signal processing system in accordance with the present invention is shown. A single channel signal input 10 is connected toA speech separator 12. The speech splitter 12 is connected to a multi-channel emulator 16 by a non-speech signal line 14 and to a post-emulation processing system 20 by a speech signal line 18. The multi-channel emulator 16 passes through the emulated signal line 22_A-22_ZConnected to a post-simulation processing system 20. The speech separator 12 has two output taps, a speech level tap 26 and a non-speech level tap 28.

In operation, a single channel signal, for example a mono audio signal, is input at input 10. The single channel input signal is separated into a speech signal and a non-speech signal by a speech separator 12. The speech signal is output on line 18 as a first output channel signal to a post-simulation processing system 20. The non-speech signal portion on line 14 is then processed by a multi-channel emulator 16 to produce a plurality of output audio channel signals which are then processed by a post-emulation processing system 20. The elements and functions of post-simulation processing system 20 are shown in more detail in fig. 3a-3d and are described in more detail in corresponding portions of the present disclosure.

The speech separator 12 may comprise a band-pass filter whose transmission band is a frequency range, for example 300Hz to 3kHz, or for example comprise a so-called "weighted" filter, as described in publication ANSIS1.4-1983, published by the american institute for Physics for the acoustic Society of America, which contains the frequency range or spectral components normally associated with speech. Other filters with different characteristics may be employed to account for different languages, tones, etc. The speech separator 12 may also comprise a more complex filter network or other type of speech recognition device, such as a microprocessor that adaptively recognizes the signal pattern representative of speech.

The audio signal processing system according to fig. 1 is advantageous because a signal source (e.g. a video cassette tape) transmitting or having a mono audio track may be present in a five channel audio system with a realistic "surround" effect, including at the screen position of a dialog.

Referring now to fig. 2a, one embodiment of a circuit implementing the voice splitter 12 and the multi-channel emulator 16 is shown. The circuit has a single input channel and five output channels. The input channel may be a mono audio signal input and the five output channels may be a left channel, a right channel, a left surround channel, a right surround channel and a center channel, as in a home cinema system.

The input channel may be a mono audio signal input and the five output channels may be a left channel, a right channel, a left surround channel, a right surround channel and a center channel, as in a home cinema system.

The speech separator 12 may comprise an input 10 connected to an input of a speech filter 80 to a + input of a first signal adder 82 and to a + input of a second signal adder 84. An output of the speech filter 80 is connected to the first multiplier 55 and the speech level tap 26 and to a-input of a first signal adder 82. The output of the first multiplier 55 is connected to the center channel signal line 22C and to an-input of a second signal adder 84. The output of the second signal summer 84 is connected to the multi-channel emulator 16 by the non-speech content signal line 14. The output of the first signal adder 82 is connected to the non-speech level tap 28.

The non-speech content signal line 14 is connected to the + input terminal of the third signal adder 34 and one terminal of the fourth signal adder 36 through the delay unit 32, thereby providing a plurality of paths for processing non-speech signals. An output of delay element 32 is coupled to a-input of fourth signal adder 36, a + input of seventh signal adder 46 and a + input of eighth signal adder 48. An output of the third signal adder 34 is coupled to an input of a fifth signal adder 38 and to an input of a second multiplier 40. An output of the fourth signal adder 36 is connected to a + input of a sixth signal adder 42 and to one input of a third multiplier 44. An output of the fifth signal adder 38 is connected to the left channel signal line 221 and to an-input of the seventh signal adder 46.

An output of the sixth signal adder 42 is connected to the right channel signal line 22R and to a + input of an eighth signal adder 48. An output terminal of the seventh signal adder 46 is connected to the right surround channel signal line 22 Rs. The output terminal of the eighth signal adder 48 is connected to the left surround signal line 22 Ls. An output of delay element 32 is coupled to an input of a seventh signal adder 46 and to an input of an eighth signal adder 48.

The delay unit 32 may delay the signal by 5 ms. The third signal adder 34 may scale the input of the delay unit 32 by a factor of 0.5. The fourth signal adder 36 may scale the input of the delay unit 32 by a factor of 0.5. Seventh signal adder 46 and eighth signal adder 48 may scale their outputs by a factor of 0.5. The first multiplier 55 may multiply the input signal from the voice filter 80 by a coefficient | C |/(| C | + | C)^-L) (hereinafter α), where | C | is the time-averaged amplitude of the speech signal on line 18 and | C |^-And | is the time-averaged amplitude of the negation of the speech signal. | C | and | C |^-L may be measured at the voice tap 26 and the non-voice tap 28, respectively. The time averaging of | C | and | C | can be performed during the sampling period, e.g., 300 ms. The time averaging of the | C | values can also be performed during two different times, e.g. 300ms and 30ms, combined and scaled.

Multipliers 40, 44 may multiply their inputs by a factor alpha.

For a monophonic input signal M, the circuit of fig. 2a gives the following output signals on the following signal lines:

table 1:

the signal value is

Line channel signal α → 0 α → 1

22C center

22L left (L)

22R Right (R)

22Ls left surround

22Rs right surround

Where C represents the speech content of the signal M, C^-Representing the non-speech content of the signal M, CD representing the non-speech content of the time delayed signal M, L representing the left channel signal, R representing the right channel signal, and α is as defined above.

Referring now to fig. 2b, a second embodiment of a circuit implementing the voice splitter 12 and the multi-channel emulator 16 is shown. The circuit includes a single input channel and five output channels. The input channel may be a mono audio input and the five output channels may be a left channel, a right channel, a left surround channel, a right surround channel and a center channel, as in a home cinema system.

The circuit of fig. 2b is essentially the same as that of fig. 2a, except that in fig. 2b the input to multiplier 55 is connected directly to input 10 rather than to the output of speech filter 80 and the signal on center channel signal line 22C is scaled by a factor of 1.414.

The circuit according to the invention is advantageous because it can provide realistic five channel effects from a mono signal. In the left and right channels, C^-The components are in phase, but the 5CD components are out of phase, which produces a stereo effect. In the left and right surround channels, C^-The components are out of phase, which prevents localization in the left surround and right surround channels. The speech content of the signal M is radiated only by the center channel and is scaled to provide the appropriate power level so that the speech sum is localized at the screen and the appropriate level.

The circuit according to the invention is also advantageous in that the total signal power is maintained. As can be seen in this circuit, if fig. 2a and 2b, and table 1, variable gain a is added directly to the signal in channel 22C and the signal at ± 5CD is subtractively combined with the signal in channels 22L and 22R, so that an increase in variable gain a results in an increase in the signal strength of the signal in channel 22C and a decrease in the signal strength of the signals in channels 22L and 22R.

The circuit according to the invention is also advantageous because the relative proportions of the sounds radiated by the loudspeakers connected to the different channels are appropriate with respect to the speech content of the monophonic input signal. If the input signal M contains non-speech, C approaches zero, C^-Toward M, and α toward zero. In this case, no signal at the center channel and signals at the other channels are shown in table 1. If the signal M is predominantly speech, C approaches M, C^-Approaching zero and alpha approaching one. In this case, the signals in the left and right surround channels approach zero, and the signals in the left and right channels approach C, respectively^-D and CD. Because the signal is delayed, the center channel is the source of the first arrival information, and the information from the complementary channels arrives later in time, so that the listener locates the radiation from the center channel. When the signal is primarily speech, the signals at the left surround and right surround channels approach zero so that there is no radiation from the surround speakers.

A further advantage of the circuit according to the invention is that the combined effect of the circuit is time-varying, so that the source of the left and right channel perceptions is not spatially fixed.

Referring to FIGS. 3a-3d, an alternative embodiment of a post-simulation processing system 20 is shown. In fig. 3a, the signal lines 22L, 22Ls, 22R, 22Rs and 22C may be connected to respective electroacoustic transducers 52L, 52Ls, 52R, 52Rs and 52C, which emit sound waves corresponding to the signals on the signal lines 22L, 22Ls, 22R, 22Rs and 22C, respectively. The electroacoustic sensors 52L, 52Ls, 52R, 52Rs, and 52C may be left, left surround, right surround, and center channel speakers of a home theater system.

In the embodiment of fig. 3b, post-simulation processing system 20 may include a crossover network 54 connecting signal lines 22L, 22Ls, 22R, and 22Rs to tweeters 56L, 56Ls, 56R, and 56Rs, respectively, and to sub-woofer 58, while signal line 22C may be connected to an electro-acoustic transducer 60.

The tweeters 56L, 56Ls, 56R, and 56Rs may be left, left surround right, and right surround speakers, the subwoofer 58 may be the subwoofer, and the electroacoustic transducer 60 may be a center channel of a subwoofer/satellite home theater system.

In the embodiment of fig. 3c, post-simulation processing system 20 may include a circuit for down-converting the output of multi-channel simulator 16 into a three-channel signal suitable for recording, transmission, or playback in a three-channel system. The input terminals of the ninth signal adder 62 are connected to the signal lines 22Ls and 22 Rs. An output of ninth signal adder 62 is coupled to an input of a tenth signal adder 64 and to an input of an eleventh signal adder 66. The signal from the ninth signal adder 62 to the tenth signal adder 64 may be scaled by a factor of 0.707 and the signal from the ninth signal adder 62 to the eleventh signal adder 66 may be scaled by a factor of-0.707. An input of the tenth signal adder 64 may be coupled to the signal line 22L such that the output signal of the tenth signal adder 64 is 0.707(Ls + Rs) + L, (where Ls, Rs, and L represent inputs from the signal lines 22Ls, 22Rs, and 22L, respectively), which is output at the left channel output 86L. An input of the eleventh signal adder 66 may be coupled to the signal line 22R such that the output signal of the eleventh signal adder 66 is-0.707 (Ls + Rs) + R, (where Ls, Rs and R represent inputs from the signal lines 22Ls, 22Rs and 22R, respectively), which is output at the right channel output 86R. Signal line 22C is connected to center channel output 86C.

In the embodiment of fig. 3d, post-simulation processing system 20 includes a circuit for down-converting the output signal of multi-channel simulator 16 to a two-channel signal suitable for recording, transmission or for playback in a two-channel system. The input terminals of the signal adder 62 are connected to the signal lines 22Ls and 22 Rs. An output of ninth signal adder 62 is coupled to an input of a tenth signal adder 64 and to an input of an eleventh signal adder 66. The signal from the ninth signal adder 62 to the tenth signal adder 64 may be scaled by a factor of 0.707 and the signal from the ninth signal adder 62 to the eleventh signal adder 66 may be scaled by a factor of-0.707. The input of the tenth signal adder 64 is connected to the signal line 22L such that the output signal of the tenth signal adder 64 is 0.707(Ls + Rs) + L, (where Ls, Rs, and L represent the signals on the signal lines 22Ls, 22Rs, and 22L, respectively). An output of the tenth signal adder 64 is connected to an input of a twelfth signal adder 68. An input terminal of the eleventh signal adder 66 may be coupled to the signal line 22R such that the output signal of the eleventh signal adder 66 is-0.707 (Ls + Rs) + R, (where Ls, Rs and R represent inputs from the signal lines 22Ls, 22Rs and 22R, respectively), which is output at the right channel output signal terminal 86R. An output of the eleventh signal adder 66 is connected to an input of a thirteenth signal adder 70. The signal from the first multiplier 55 to the tenth signal adder 68 may be scaled by a factor of 0.707, so that the output signal of the tenth signal adder 68 is 0.707C +707(Ls + Rs) + L, (where Ls, Rs, L and C represent the inputs from the signal lines 22Ls, 22Rs and 22L and from the first multiplier 55, respectively). The output of the tenth signal adder is connected to the left channel termination output 84L. The signal from the first multiplier 55 to the thirteenth signal adder 70 may be scaled by a factor of 0.707, so that the output signal of the thirteenth signal adder 70 is 0.707C-707(Ls + Rs) + L, (where Ls, Rs, L and C represent inputs from signal lines 22Ls, 22Rs and 22L and 22C, respectively). An output of the thirteenth signal adder 70 is connected to the right channel output 84R.

The embodiments of fig. 3c and 3d are advantageous because they can be re-recorded or re-broadcast in a two or three channel format and subsequently decoded to be provided in a five channel format.

Referring now to fig. 4, a circuit embodying the principles of the present invention in a two input channel system is shown. Input channel terminal 90L is connected to an input of left speech filter 92L and additively connected to left adder 94L, and an output of speech filter 92L is differentially connected to an input of left adder 94L and additively connected to center adder 96C.

The output of the left adder 94L is connected to the left channel output 98L and the left surround adder 94Ls, and differentially connected to the right surround adder 94 Rs. The right input channel terminal 90R is connected to the input of a right speech filter 92L and additively connected to a right adder 94R. The output of speech filter 92R. The output of speech filter 92R is connected to the input of the right adder and additively connected to the center adder 96C. The output of the right adder 94R is connected to the right channel output 98R and the right surround adder 94Rs, and is differentially connected to the left surround adder 94 Ls. The output of the left surround adder 94Ls is connected to a left surround output 98Ls, and the output of the right surround adder 94Rs is connected to a right surround output 98 Rs.

In operation, a two-channel input signal, such as a stereo signal having left and right channels, is input at input terminals 90L and 90R, respectively. The circuit separates the voiceband portions of the signal and combines the left voiceband portion C_LAnd a right voiceband portion C_RCombine them and scale them to form the center channel signal output at the center channel terminal 98C. The non-speech portion of the left channel signal and the non-speech portion of the right channel signal are output at a left channel output 98L and a right channel output 98R, respectively. The output of center channel terminal 98C may then be used as the center channel for a three or five channel audio system. The outputs of the left channel output 98L and the right channel output 98R can be used as the left and right channels of a three-channel system. If a five channel output is desired, the output of the adder 94R may be differentially combined with the output of the adder 94L and scaled to form a left surround channel signal that is output at a left surround output 98Ls, while the output of the adder 94L may be differentially combined with the output of the adder 94R and scaled to form a right surround channel signal that is output at a right surround output 98 Rs.

Other embodiments are within the claims.

Claims

1. A method for processing a single channel audio signal to provide a plurality of audio channel signals, comprising:

separating the single-channel audio signal into a first separated signal and a second separated signal characterized by a spectral pattern general characteristic of speech;

processing the first split signal to provide a first audio-channel signal; and

processing the second split signal to generate a remaining channel signal of the plurality of channel signals.

2. The method of processing an audio signal according to claim 1, wherein said modifying comprises:

splitting the second split signal into a plurality of signals;

and one of the latter signals is multiplied by a predetermined coefficient.

3. A method of processing an audio signal according to claim 2, wherein the coefficients are variable with respect to time.

4. A method of processing an audio signal according to claim 2, wherein said coefficient applies a gain proportional to the sum of the time-averaged amplitude of said first split signal and the time-averaged amplitude of said second split signal divided by the time-averaged amplitude of said first split signal.

5. The method of processing an audio signal according to claim 1, wherein said modifying comprises dividing said second split signal into a plurality of signals; and

time-delaying the second split signal.

6. A method of processing an audio signal according to claim 1, wherein said modifying step provides a left channel signal and a right channel signal.

7. A method of processing an audio signal according to claim 6, wherein said modifying step additionally provides a left surround channel signal and a right surround channel signal.

8. A method of processing an audio signal according to claim 1, wherein said first audio channel signal is a center channel signal.

9. The method of processing a single channel audio signal according to claim 8, wherein said processing said first split signal comprises multiplying said first split signal by a first predetermined coefficient.

10. A method of processing a single audio signal according to claim 9, wherein said modifying step comprises the step of multiplying said second split signal by a second predetermined coefficient.

11. The method of processing a single audio signal according to claim 10, wherein said first predetermined coefficient and said second predetermined coefficient are determined by: an increase in signal strength of the first split signal coincides with a decrease in signal strength of the second split signal.

12. The method of processing a single channel audio signal according to claim 9, wherein said first predetermined coefficient is variable with respect to time.

13. The method of processing a single channel audio signal according to claim 9, wherein the predetermined coefficient is proportional to a total of a time average of the first split signal and a time average amplitude of the second split signal divided by the time average amplitude of the first split signal.

14. An audio signal processing apparatus for processing a single channel audio signal to provide a plurality of audio channel signals, comprising

A separator for separating the audio signal into a first separated signal and a second separated signal characterized by spectral general characteristics of speech; and

a first circuit, coupled to said splitter and said speech splitter, for providing a first subset of said plurality of audio channel signals, said splitter responsive to said second split signal.

15. The audio signal processing apparatus of claim 14, wherein the first circuit comprises a plurality of signal paths for the second split signal,

one of the plurality of signal paths is provided with a time delay.

16. The audio signal processing apparatus of claim 14, wherein the first circuit comprises a plurality of signal paths,

at least one of the plurality of signal paths includes a multiplier.

17. The audio signal processing apparatus of claim 16 wherein said first plurality of signal paths are constructed and arranged to subtractively combine signals to which said variable gain has been added with signals to which said variable gain has not been added.

18. The audio signal processing apparatus of claim 14, wherein the first subset of the plurality of audio channel signals comprises a left channel signal and a right channel signal.

19. The audio signal processing apparatus of claim 18, wherein the first subset of the plurality of audio channel signals comprises a left surround channel signal and a right surround channel signal.

20. The audio signal processing apparatus of claim 14, wherein the separator comprises a band-pass filter having a pass band substantially corresponding to a characteristic band of a speech spectrum.

21. The audio signal processing apparatus of claim 14 further comprising a second circuit coupled to said splitter and responsive to said first split signal for providing a second subset of said plurality of audio channel signals.

22. The audio signal processing device of claim 21, wherein the second subset comprises a single audio channel signal.

23. The audio signal processing apparatus of claim 22, wherein the mono audio channel signal is a center channel signal.

24. An audio signal processing system comprising;

an input for a single input channel signal;

a center channel output for a center channel output signal C;

a plurality of further outputs for a corresponding plurality of further output audio channel signals;

a separator for separating said single channel input signal into a speech audio signal and a non-speech audio signal;

a first circuit for connecting said voice audio signal to said center channel terminal, and

a second circuit connecting the splitter to the plurality of outputs provides a corresponding plurality of other audio channel signals in response to the non-speech signal.

25. An audio signal processing system according to claim 24, wherein said second circuit comprises a plurality of signal paths, one of said plurality of signal paths being provided with a time delay.

26. An audio signal processing system according to claim 24, wherein said circuit comprises a plurality of signal paths, at least one signal path of said plurality of signal paths comprising a multiplier.

27. An audio signal processing system according to claim 26, wherein said multiplier is connected to the other output, the left channel output.

28. An audio signal processing system according to claim 26, wherein the multiplier is connected to the other output, the right channel output.

29. An audio signal processing system as claimed in claim 24, wherein the splitter comprises a band pass filter having a pass band substantially corresponding to the spectrum of the speech signal.

30. An audio signal processing system according to claim 24, further comprising a multiplier connecting said splitter to said center channel output and the output of said splitter by a predetermined factor.

31. The audio signal processing system of claim 30, wherein the predetermined coefficient is variable with respect to time.

32. An audio signal processing system according to claim 30, wherein the predetermined coefficient is proportional to a time-averaged amplitude of the speech audio signal.

33. An audio signal processing system according to claim 32, wherein the predetermined coefficient is proportional to the sum of the time-averaged amplitude of the speech audio signal and the time-averaged amplitude of the non-speech audio signal divided by the time-averaged amplitude of the speech audio signal.

34. The audio signal processing system of claim 24, wherein said second circuit provides a left channel signal t and a right channel signal R, a left surround channel signal Ls and a right surround channel signal Rs,

further comprising a down-conversion circuit coupled to said plurality of other outputs and said center channel output for down-converting said plurality of other output audio channel signals and said center channel signal to provide a plurality of decodable audio channel signals.

35. The audio signal processing apparatus of claim 34, where the plurality of decodable audio channel signals comprises two decodable audio channel signals.

36. The audio signal processing apparatus of claim 34, where the plurality of decodable audio channel signals comprises three decodable audio channel signals.

37. A method for processing a single channel audio signal to provide two decodable audio channel signals, a decodable audio channel signal being decodable into five audio channel signals, comprising:

processing said first split signal to provide a center channel signal C;

processing said second split signal to provide a left channel signal L, a right channel signal R, a left surround channel signal Ls and a right surround channel signal Rs;

combining the center channel signal, the sum of the left surround and the right surround channel signals and the left channel to produce a first decodable audio channel signal of the two decodable audio channel signals; and

combining the center audio channel signal, the sum of the left surround and the right surround channel signals and the right audio channel to produce a second one of the two decodable audio channel signals.

38. The method of processing a single channel audio signal according to claim 37, further comprising scaling said center channel signal and said sum of said left surround and said right surround channel signals by center and surround coefficients, respectively.

39. The method of processing a single channel audio signal according to claim 38, further comprising inverting the phase of the sum component comprising one of the first and second decodable audio signals relative to a sum component comprising the other decodable audio signal.

40. A method for processing a single channel audio signal to provide three decodable audio channel signals, the decodable audio channel signals being subsequently decodable into five audio channel signals, comprising:

processing the pair formed a center channel signal to form a center channel signal comprising a first decodable audio signal;

processing said second split signal to provide a left channel signal, a right channel signal, a left surround channel signal and a right surround channel signal;

combining the sum of the left surround and the right surround channel signals with the left audio channel to produce a first of the two decodable audio channel signals; and

combining a sum of said left surround and said right surround channel signals with a right channel signal to produce a third of said decodable audio channel signals.

41. The method of processing a single channel audio signal according to claim 40, further comprising scaling with a predetermined surround factor.

42. The method of processing a single channel audio signal according to claim 41, further comprising inverting the phase of the sum component, the sum comprising one of the second and third decodable audio signals relative to the other and the second and third decodable audio signals.

43. A method of processing two input audio channel signals to provide two or more output audio channel signals, comprising:

separating each of the two input audio channel signals into a first separated signal and a second separated signal characterized by a general characteristic of a spectral pattern of speech;

combining a first split signal of the first input audio channel signal with a first split signal of the second input audio channel signal to form a first one of the two or more output audio channel signals;

the second split signal of the first input signal comprises a second output audio channel signal of the two or more output audio channel signals; and

the second split signal of the second input signal comprises a third output channel signal of the two or more output channel signals.

44. The method of processing two input audio channel signals according to claim 43, wherein said second split of said first input signal comprises providing a left channel signal and said second split of said second input signal comprises a right channel signal.

45. The method of processing two input audio channel signals according to claim 43, wherein said first one of said more than two output audio channel signals comprises a center channel signal.

46. The method of processing two input audio channel signals according to claim 43, further comprising differentially combining said second split signal of said first input signal and said second split signal of said second input signal to form a fourth output audio channel signal of said more than two output audio channel signals; and

differentially combining the second split signal of the second input signal with the second split signal of the first input signal to form a fifth output audio channel signal of the two or more output audio channel signals.

47. An audio signal processing apparatus for processing a single channel of two audio channel signals to provide more than two output audio channel signals, comprising: a first splitter for splitting a first audio channel signal of the two audio channel signals into a first split signal characterized by a spectral pattern-generic characteristic of speech and a first output audio channel signal comprising the two or more output audio channel signals;

a second splitter for splitting a second audio channel signal of the two audio channel signals into a first split signal characterized by a spectral pattern characteristic of speech and a second split signal comprising a second output audio channel signal of the two or more output audio channel signals; and

a first combiner for combining the first split signal of the first audio channel signal and the first split signal of the second audio channel signal to provide a third output audio channel signal of the two or more output audio channel signals.

48. The audio signal processing device of claim 47, further comprising

A second combiner for differentially combining the first output audio channel signal and the second output channel signal to provide a fourth output audio channel of the two or more output audio channels; and

a third combiner for differentially combining the second output audio channel signal with the first output audio channel to provide a fifth output audio channel of the two or more output audio channels.