Audio signal decorrelator, multi channel audio signal processor, audio signal processor, method for deriving an output audio signal from an input audio signal and computer program

• US 8,538,037 B2
• Filed: 03/28/2007
• Issued: 09/17/2013
• Est. Priority Date: 04/13/2006
• Status: Active Grant

• Current Assignee: Fraunhofer Gesellschaft Zur Foerderung Der Angewandten Forsching E.V.
• Original Assignee: Fraunhofer Gesellschaft Zur Foerderung Der Angewandten Forsching E.V.
• Inventors: Buchner, Herbert, Herre, Jrgen
• Primary Examiner(s): Luke, Daniel
• Assistant Examiner(s): Ahmad, Khaja
• Application Number:
  US12/296,698
• Publication Number:
  US 20090304198A1
• Time in Patent Office: 2,365 Days
• Field of Search: 381/66, 381/93, 381/94.1, 381/94.3, 37940601-40616, 370/289, 370282-286, 455/570
• US Class Current: 381/94.3
• CPC Class Codes: G10L 19/008 Multichannel audio signal coding or decoding using interchannel correlation to reduce redundancy, e.g. joint-stereo, intensity-coding or matrixing
• CPC Class Codes: G10L 2021/02082 the noise being echo, reverberation of the speech
• CPC Class Codes: H04M 9/082 using echo cancellers echo cancellers per se H04B3/23

Abstract

An audio signal decorrelator for deriving an output audio signal from an input audio signal has a frequency analyzer for extracting from the input audio signal a first partial signal descriptive of an audio content in a first audio frequency range and a second partial signal descriptive of an audio content in a second audio frequency range having higher frequencies compared to the second audio frequency range. A partial signal modifier modifies the first and second partial signals, to obtain first and second processed partial signals, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than that applied to the second partial signal, or for modifying only the first partial signal. A signal combiner combines the first and second processed partial signals, or combines the first processed partial signal and the second partial signal, to obtain an output audio signal.

Claims

1. An audio signal decorrelator for deriving an output audio signal from an input audio signal, the audio signal decorrelator comprising:

a frequency analyzer for extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

a modulation function generator to produce a time-variant modulation function;

a partial signal modifier adapted to time-variantly modify the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or to time-variantly modify only the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

a signal combiner for combining the first processed partial signal and the second processed partial signal, or for combining the first processed partial signal and the second partial signal, to acquire an output audio signal;

wherein

the frequency analyzer comprises an analysis filterbank adapted to decompose the input audio signal into at least two spectral coefficients, a first spectral coefficient being descriptive of the audio content in the first audio frequency range, the first spectral coefficient forming the first partial signal, a second spectral coefficient being descriptive of the audio content in the second audio frequency range, the second spectral coefficient forming the second partial signal;

the partial signal modifier is adapted to modulate a phase at least of the first spectral coefficient, to acquire the first processed partial signal, or to modulate the phase of the first spectral coefficient, to acquire the first processed partial signal, and a phase of the second spectral coefficient, to acquire a second processed partial signal; and

the signal combiner comprises a synthesis filterbank adapted to combine the processed first partial signal and the second partial signal to acquire the output audio signal as a time domain signal, or to combine the processed first partial signal and the processed second partial signal to acquire the output audio signal as a time domain signal.

2. The audio signal decorrelator according to claim 1, wherein the frequency analyzer is adapted to decompose temporal blocks of the input audio signal into a plurality of spectral coefficients, such that the first partial signal comprises a time sequence of first spectral coefficients, and such that the second partial signal comprises a time sequence of second spectral coefficients;

wherein the partial signal modifier is adapted to process at least the first partial signal to acquire the first processed partial signal as a time sequence of processed first spectral coefficients, and

wherein the synthesis filterbank is adapted to combine the time sequence of spectral coefficients of the first processed partial signal and the time sequence of spectral coefficients of the second partial signal or the second processed partial signal into the time domain output audio signal.

3. The audio signal decorrelator according to claim 1, wherein the frequency analyzer is adapted to provide the spectral coefficients as complex values;

wherein the partial signal modifier is adapted to time-variantly modulate a phase of the first spectral coefficient or of the first spectral coefficients by multiplying or dividing the first spectral coefficient or the first spectral coefficients with a first plurality of time variant complex values (ejφ

1) comprising a predetermined, identical, time invariant magnitude and a time variant phase.

4. The audio signal decorrelator according to claim 3,

wherein the audio signal decorrelator is adapted to derive the time variant phase of the complex values of the first plurality of time variant complex values from values of the modulation function at different instances of time.

5. The audio signal decorrelator according to claim 4, wherein the modulation function generator is adapted to produce a sinusoidal function of time as the modulation function.

6. The audio signal decorrelator according to claim 4, wherein the modulation function generator is adapted to produce the modulation function in such a way that a modulation frequency introduced into the audio content represented by the first partial signal due to a time variation of the phase of the time variant complex values of the first plurality of time variant complex values is smaller than 10 Hz.

7. The audio signal decorrelator according to claim 4, wherein the modulation function generator is adapted to produce a periodic modulation time function comprising a frequency of less than 10 Hz.

8. The audio signal decorrelator according to claim 4, wherein the modulation function generator is adapted to produce a random or pseudo-random function of time.

9. The audio signal decorrelator according to claim 4, wherein the partial signal modifier is adapted to modulate phases of the spectral coefficients of the second partial signal by multiplying or dividing the second spectral coefficient or the second spectral coefficients with a second plurality of time variant complex values comprising a predetermined, identical, time invariant magnitude and a time-variant phase;

and

wherein the audio signal decorrelator is adapted to derive time variant phases of the complex values of the second plurality of time variant complex values from values of the modulation function at different instances of time.

10. The audio signal decorrelator according to claim 9, further comprising a scaler adapted to scale the modulation time function with a first time-invariant scalar value to acquire the time variant phase of the complex values of the first plurality of time variant complex values, and to scale the modulation time function with a second time-invariant scalar value to acquire the time variant phase of the complex values of the second plurality of the time variant complex values;

wherein an absolute value or a magnitude of the first scalar value is larger than an absolute value or a magnitude of the second scalar value, such that a phase shift modulation amplitude applied to the first partial signal is larger than a phase shift modulation amplitude applied to the second partial signal.

11. The audio signal decorrelator according to claim 1, wherein the frequency analyzer is adapted to extract a third partial signal from the input audio signal, the third partial signal being descriptive of an audio content in a third audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range, and the second audio frequency range comprising higher frequencies compared to the third audio frequency range;

wherein the partial signal modifier is adapted to modify the first partial signal, to acquire a first processed partial signal, to modify the second partial signal, to acquire a second processed partial signal, and to modify the third partial signal, to acquire a third processed partial signal, so that a time variant phase shift modulation is applied to the first partial signal which comprises a larger modulation amplitude than a time variant phase shift modulation applied to the second partial signal, wherein a modulation amplitude of the time-variant phase shift applied to the second partial signal is larger than a modulation amplitude of a time variant phase shift applied to the third partial signal; and

wherein the signal combiner is adapted to combine the first processed partial signal, the second processed partial signal and the third processed partial signal, to acquire the output audio signal.

12. The audio signal decorrelator according to claim 1, wherein a lower frequency border of the first audio frequency range is larger than 1.1 kHz.

13. A multi channel audio signal processor for decorrelating a first channel audio signal and a second channel audio signal, to acquire a first decorrelated output audio signal and a second decorrelated output audio signal, the multi channel audio signal processor comprising:

a first audio signal decorrelator for deriving an output audio signal from an input audio signal, the audio signal decorrelator comprising;

a frequency analyzer for extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

a modulation function generator to produce a time-variant modulation function;

a partial signal modifier adapted to time-variantly modify the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or to time-variantly modify only the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

a signal combiner for combining the first processed partial signal and the second processed partial signal, or for combining the first processed partial signal and the second partial signal, to acquire an output audio signal, wherein the first audio signal decorrelator is adapted to receive as the input audio signal the first channel audio signal, and wherein the output audio signal of the first audio signal decorrelator forms the first decorrelated output audio signal;

wherein

the second channel audio signal or a signal derived therefrom forms the second decorrelated output audio signal;

the first audio signal decorrelator is adapted such that audio contents in the first audio frequency range of the first decorrelated output audio signal and of the second decorrelated output audio signal comprise a stronger time-variant phase shift modulation with respect to each other than audio contents in the second audio frequency range of the first decorrelated output audio signal and of the second decorrelated audio output signal;

the frequency analyzer comprises an analysis filterbank adapted to decompose the input audio signal into at least two spectral coefficients, a first spectral coefficient being descriptive of the audio content in the first audio frequency range, the first spectral coefficient forming the first partial signal, a second spectral coefficient being descriptive of the audio content in the second audio frequency range, the second spectral coefficient forming the second partial signal;

the partial signal modifier is adapted to modulate a phase at least of the first spectral coefficient, to acquire the first processed partial signal, or to modulate the phase of the first spectral coefficient, to acquire the first processed partial signal, and a phase of the second spectral coefficient, to acquire a second processed partial signal; and

the signal combiner comprises a synthesis filterbank adapted to combine the processed first partial signal and the second partial signal to acquire the output audio signal as a time domain signal, or to combine the processed first partial signal and the processed second partial signal to acquire the output audio signal as a time domain signal.

14. The multi channel audio signal processor according to claim 13, further comprising a second audio signal decorrelator, the second audio signal decorrelator comprising:

a frequency analyzer for extracting a third partial signal and a fourth partial signal from the second channel audio signal, the third partial signal being descriptive of an audio content in a third audio frequency range, the fourth partial signal being descriptive of an audio content in a fourth audio frequency range, the third audio frequency range comprising higher frequencies than the fourth audio frequency range;

a partial signal modifier for modifying the third partial signal to acquire a third processed partial signal, so that a modulation amplitude of a time variant phase shift or a time variant delay applied to the third partial signal is larger than a modulation amplitude of a time variant phase shift or a time variant delay applied to the fourth partial signal, or for modifying only the third partial signal, so that a time variant phase shift or a time variant delay is applied to the third partial signal; and

a signal combiner for combining the third processed partial signal and the fourth processed partial signal, to acquire the second decorrelated output signal, or for combining the third processed partial signal and the fourth partial signal, to acquire the second decorrelated output signal.

15. The audio signal processor according to claim 13, further comprising a second audio signal decorrelator for deriving an output audio signal from an input audio signal, the audio signal decorrelator comprising:

a frequency analyzer for extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

a partial signal modifier for modifying the first partial signal, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or for modifying only the first partial signal, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

a signal combiner for combining the first processed partial signal and the second processed partial signal, or for combining the first processed partial signal and the second partial signal, to acquire an output audio signal,

wherein the first audio frequency range of the first audio signal decorrelator comprises higher frequencies than the second audio frequency range of the second audio signal decorrelator,

wherein the first audio frequency range of the first audio signal decorrelator overlaps with the first audio frequency range of the second audio signal decorrelator in a first overlap audio frequency range,

wherein the second audio frequency range of the first audio signal decorrelator overlaps with the second audio frequency range of the second audio signal decorrelator in a second overlap audio frequency range;

wherein the second audio signal decorrelator is adapted to receive as the input audio signal the second channel input audio signal;

wherein the output audio signal of the second audio signal decorrelator forms the decorrelated second output audio signal;

wherein the first audio signal decorrelator and the second audio signal decorrelator are adapted such that the time variant phase shift or time variant delay introduced into the first partial signal of the first audio signal decorrelator is different from the time variant phase shift or time variant delay introduced into the first partial signal of the second audio signal decorrelator.

16. The multi channel audio signal processor according to claim 15, wherein the first signal decorrelator is adapted to adjust the time variant delay or time variant phase shift in its first audio frequency range based on a first phase shift value, and to adjust the time variant delay or time variant phase shift in its second audio frequency range based on a second phase shift value;

wherein the second audio signal decorrelator is adapted to adjust the time variant delay or the time variant phase shift in its first audio frequency range based on a third phase shift value, and to adjust the time variant delay or time variant phase shift in its second audio frequency range based on a fourth phase shift value; and

wherein the multi channel audio signal processor further comprises a scaler adapted to derive the first phase shift value, the second phase shift value, the third phase shift value and the fourth phase shift value from a common modulation time function by scaling the modulation time function, to acquire scaled values, and deriving the phase shift values from the scaled values.

17. The multi channel audio signal processor according to claim 16, wherein the first audio signal decorrelator, the second audio signal decorrelator and the scaler are adapted such that a phase shift applied in the first overlap audio frequency range by the first audio signal decorrelator comprises an opposite direction when compared to a phase shift applied in the first overlap audio frequency range by the second audio signal decorrelator, and that a phase shift applied in the second overlap audio frequency range by the first audio signal decorrelator comprises an opposite direction when compared to a phase shift applied in the second overlap audio frequency range by the second audio signal decorrelator.

18. The multi channel audio signal processor according to claim 16, wherein the first audio signal decorrelator, the second audio signal decorrelator and the scaler are adapted such that a phase shift applied in the first overlap audio frequency range by the first audio signal decorrelator comprises the same magnitude as a phase shift applied in the first overlap audio frequency range by the second audio signal decorrelator, and

that a phase shift applied in the second overlap audio frequency range by the first audio signal decorrelator comprises the same magnitude as a phase shift applied in the second overlap audio frequency range by the second audio signal decorrelator.

19. The multi channel audio signal processor according to claim 15, wherein the first audio signal decorrelator comprises an analysis filterbank adapted to decompose the first channel audio signal into at least two spectral coefficients, a first spectral coefficient being descriptive of the audio content in the first audio frequency range, the first spectral coefficient forming the first partial signal, a second spectral coefficient being descriptive of the audio content in the second audio frequency range, the second spectral coefficient forming the second partial signal;

wherein the second audio signal decorrelator comprises an analysis filterbank adapted to decompose the second channel audio signal into at least two spectral coefficients, a third spectral coefficient being descriptive of the audio content in the second audio frequency range, the third spectral coefficient forming the first partial signal of the second audio signal decorrelator, a fourth spectral coefficient being descriptive of the audio content in the second audio frequency range, the fourth spectral coefficient forming the second partial signal of the second audio signal decorrelator,

wherein the multi channel audio signal processor, the first audio signal decorrelator and the second audio signal decorrelator are adapted to achieve that in a given time interval, the first spectral coefficient and the third spectral coefficient are multiplied with values being complex conjugate with respect to each other, or to achieve that in the given time interval, the first spectral coefficient is multiplied with a complex value of magnitude 1, while the third spectral coefficient is divided by the same complex value.

20. An audio signal processor for receiving a first channel audio signal, a second channel audio signal, a third channel audio signal, a fourth channel audio signal and a fifth channel audio signal being part of a multi channel audio signal, the audio signal processor comprising:

a first multi channel audio signal processor for decorrelating a first channel audio signal and a second channel audio signal, to acquire a first decorrelated output audio signal and a second decorrelated output audio signal, the multi channel audio signal processor comprising;

a first audio signal decorrelator for deriving an output audio signal from an input audio signal, the audio signal decorrelator comprising;

a frequency analyzer for extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

a modulation function generator to produce a time-variant modulation function;

a partial signal modifier adapted to time-variantly modify the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or to time-variantly modify only the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

a signal combiner for combining the first processed partial signal and the second processed partial signal, or for combining the first processed partial signal and the second partial signal, to acquire an output audio signal, wherein the first audio signal decorrelator is adapted to receive as the input audio signal the first channel audio signal, and wherein the output audio signal of the first audio signal decorrelator forms the first decorrelated output audio signal;

wherein

the second channel audio signal or a signal derived therefrom forms the second decorrelated output audio signal; and

the first audio signal decorrelator is adapted such that audio contents in the first audio frequency range of the first decorrelated output audio signal and of the second decorrelated output audio signal comprise a stronger time-variant phase shift modulation with respect to each other than audio contents in the second audio frequency range of the first decorrelated output audio signal and of the second decorrelated audio output signal, wherein the first multi channel audio signal processor receives as its first channel audio signal the first channel audio signal of the multi channel signal, and wherein the first multi channel audio signal processor receives as its second channel audio signal the second channel audio signal of the multi channel signal;

a second multi channel audio signal processor for decorrelating a first channel audio signal and a second channel audio signal, to acquire a first decorrelated output audio signal and a second decorrelated output audio signal, the multi channel audio signal processor comprising;

a first audio signal decorrelator for deriving an output audio signal from an input audio signal, the audio signal decorrelator comprising;

a frequency analyzer for extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

a partial signal modifier adapted to time-variantly modify the first partial signal, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or to time-variantly modify only the first partial signal, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

a signal combiner for combining the first processed partial signal and the second processed partial signal, or for combining the first processed partial signal and the second partial signal, to acquire an output audio signal, wherein the first audio signal decorrelator is adapted to receive as the input audio signal the first channel audio signal, and wherein the output audio signal of the first audio signal decorrelator forms the first decorrelated output audio signal;

the second channel audio signal or a signal derived therefrom forms the second decorrelated output audio signal; and

the first audio signal decorrelator is adapted such that audio contents in the first audio frequency range of the first decorrelated output audio signal and of the second decorrelated output audio signal comprise a stronger time-variant phase shift modulation with respect to each other than audio contents in the second audio frequency range of the first decorrelated output audio signal and of the second decorrelated audio output signal, wherein the second multi channel audio signal processor receives as its first channel audio signal the third channel audio signal of the multi channel signal, and wherein the second multi channel audio signal processor receives as its second channel audio signal the fourth channel audio signal of the multi channel signal; and

an audio signal decorrelator for deriving an output audio signal from an input audio signal, the audio signal decorrelator comprising;

a frequency analyzer for extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

a partial signal modifier for modifying the first partial signal, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or for modifying only the first partial signal, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

a signal combiner for combining the first processed partial signal and the second processed partial signal, or for combining the first processed partial signal and the second partial signal, to acquire an output audio signal, wherein the audio signal decorrelator receives as its input audio signal the fifth channel audio signal of the multi channel signal,

the output signals of the first multi channel audio signal processor, the second multi channel audio signal processor and the audio signal decorrelator forming a decorrelated five channel signal;

the frequency analyzer comprises an analysis filterbank adapted to decompose the input audio signal into at least two spectral coefficients, a first spectral coefficient being descriptive of the audio content in the first audio frequency range, the first spectral coefficient forming the first partial signal, a second spectral coefficient being descriptive of the audio content in the second audio frequency range, the second spectral coefficient forming the second partial signal;

the partial signal modifier is adapted to modulate a phase at least of the first spectral coefficient, to acquire the first processed partial signal, or to modulate the phase of the first spectral coefficient, to acquire the first processed partial signal, and a phase of the second spectral coefficient, to acquire a second processed partial signal; and

the signal combiner comprises a synthesis filterbank adapted to combine the processed first partial signal and the second partial signal to acquire the output audio signal as a time domain signal, or to combine the processed first partial signal and the processed second partial signal to acquire the output audio signal as a time domain signal.

21. The audio signal processor according to claim 20, wherein the first channel audio signal of the multi channel signal is a left front speaker signal;

wherein the second channel audio signal of the multi channel signal is a right front speaker signal;

wherein the third channel audio signal of the multi channel signal is a left rear speaker signal;

wherein the fourth channel audio signal of the multi channel signal is a right rear speaker signal; and

wherein the fifth channel audio signal of the multi channel signal is a center speaker signal.

22. The audio signal processor according to claim 20, wherein the first multi channel audio signal processor is adapted to apply the delays in its audio frequency range in dependence on phase shift values derived from a first modulation time function using scaling operations performed by a scaler;

wherein the second multi channel audio signal processor is adapted to apply the delays in its audio frequency ranges in dependence on phase shift values derived from a second modulation time function using scaling operations performed by a scaler; and

wherein the audio signal decorrelator is adapted to apply the delays in its audio frequency ranges in dependence on phase shift values derived from a third modulation time function using scaling operations.

23. The audio signal processor according to claim 22, wherein the first modulation time function, the second modulation time function and the third modulation time function are periodic functions, and wherein the first modulation time function, the second modulation time function and the third modulation time function comprise different frequencies.

24. An audio conference system, comprising:

an audio input for receiving a first channel audio signal and a second channel audio signal;

a multi channel audio signal processor for decorrelating a first channel audio signal and a second channel audio signal, to acquire a first decorrelated output audio signal and a second decorrelated output audio signal, the multi channel audio signal processor comprising;

a first audio signal decorrelator for deriving an output audio signal from an input audio signal, the audio signal decorrelator comprising;

a frequency analyzer for extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

a modulation function generator to produce a time-variant modulation function;

a partial signal modifier adapted to time-variantly modify the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or to time-variantly modify only the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

a signal combiner for combining the first processed partial signal and the second processed partial signal, or for combining the first processed partial signal and the second partial signal, to acquire an output audio signal, wherein the first audio signal decorrelator is adapted to receive as the input audio signal the first channel audio signal, and wherein the output audio signal of the first audio signal decorrelator forms the first decorrelated output audio signal;

wherein

the second channel audio signal or a signal derived therefrom forms the second decorrelated output audio signal; and

the first audio signal decorrelator is adapted such that audio contents in the first audio frequency range of the first decorrelated output audio signal and of the second decorrelated output audio signal comprise a stronger time-variant phase shift modulation with respect to each other than audio contents in the second audio frequency range of the first decorrelated output audio signal and of the second decorrelated audio output signal, the multi channel audio signal processor receiving the first channel audio signal and the second channel audio signal;

a first electroacoustic transducer for outputting the first decorrelated output audio signal provided by the multi channel audio signal processor;

a second electroacoustic transducer for outputting the second decorrelated output audio signal provided by the multi channel audio signal processor;

a microphone acoustically coupled with a first electroacoustic transducer an a second electroacoustic transducer, the microphone providing a microphone signal descriptive of the acoustic signal;

an echo cancellation unit for reducing or removing an audio content output by the first electroacoustic transducer from the microphone signal, and for reducing or removing an audio content output by the second electroacoustic transducer from the microphone signal,

the echo cancellation unit is adapted to filter the decorellated output audio signals in dependence on filter coefficients,

to remove or subtract the filtered version of the decorrelated output audio signals from the microphone signal, to acquire an echo-reduced microphone signal, and

to adjust the filter parameters in dependence on the echo-reduced microphone signal, to minimize an audio content output by the electroacoustic transducers in the echo-reduced microphone signal;

the frequency analyzer comprises an analysis filterbank adapted to decompose the input audio signal into at least two spectral coefficients, a first spectral coefficient being descriptive of the audio content in the first audio frequency range, the first spectral coefficient forming the first partial signal, a second spectral coefficient being descriptive of the audio content in the second audio frequency range, the second spectral coefficient forming the second partial signal;

the partial signal modifier is adapted to modulate a phase at least of the first spectral coefficient, to acquire the first processed partial signal, or to modulate the phase of the first spectral coefficient, to acquire the first processed partial signal, and a phase of the second spectral coefficient, to acquire a second processed partial signal; and

the signal combiner comprises a synthesis filterbank adapted to combine the processed first partial signal and the second partial signal to acquire the output audio signal as a time domain signal, or to combine the processed first partial signal and the processed second partial signal to acquire the output audio signal as a time domain signal.

25. A method for deriving an output audio signal from an input audio signal, the method comprising:

extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

time-variantly modifying the first partial signal in dependence on a time-variant modulation function produced by a modulation function generator, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or time-variantly modifying only the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

combining the first processed partial signal and the second processed partial signal, or combining the first processed partial signal and the second partial signal, to acquire an output audio signal;

wherein

the extracting step uses an analysis filterbank adapted to decompose the input audio signal into at least two spectral coefficients, a first spectral coefficient being descriptive of the audio content in the first audio frequency range, the first spectral coefficient forming the first partial signal, a second spectral coefficient being descriptive of the audio content in the second audio frequency range, the second spectral coefficient forming the second partial signal;

the time-variantly modifying step includes modulating a phase at least of the first spectral coefficient, to acquire the first processed partial signal, or to modulate the phase of the first spectral coefficient, to acquire the first processed partial signal, and a phase of the second spectral coefficient, to acquire a second processed partial signal; and

the combining step uses a synthesis filterbank adapted to combine the processed first partial signal and the second partial signal to acquire the output audio signal as a time domain signal, or to combine the processed first partial signal and the processed second partial signal to acquire the output audio signal as a time domain signal.

28. A non-transitory computer readable medium storing a computer program comprising program code for performing the method in accordance with claim 25 when the computer program runs on a computer.

26. A method for decorrelating a first channel audio signal and a second channel audio signal, to acquire a first decorrelated output audio signal and a second decorrelated output audio signal, the method comprising:

deriving an output audio signal from an input audio signal using a method for deriving an output audio signal from an input audio signal, the method comprising;

extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

time-variantly modifying the first partial signal in dependence on a time-variant modulation function produced by a modulation function generator, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or time-variantly modifying only the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

combining the first processed partial signal and the second processed partial signal, or combining the first processed partial signal and the second partial signal, to acquire an output audio signal, wherein the first channel audio signal forms the input audio signal, and wherein the output audio signal forms the first decorrelated output audio signal; and

providing the second channel audio signal or a signal derived therefrom as the second decorrelated output audio signal;

wherein

the method for decorrelating is executed such that audio contents in the first audio frequency range of the first decorrelated output audio signal and of the second decorrelated output audio signal comprise a larger time-variant phase shift modulation amplitude with respect to each other than audio contents in the second audio frequency range of the first decorrelated output audio signal and of the second decorellated audio output signal;

the extracting step uses an analysis filterbank adapted to decompose the input audio signal into at least two spectral coefficients, a first spectral coefficient being descriptive of the audio content in the first audio frequency range, the first spectral coefficient forming the first partial signal, a second spectral coefficient being descriptive of the audio content in the second audio frequency range, the second spectral coefficient forming the second partial signal;

the time-variantly modifying step includes modulating a phase at least of the first spectral coefficient, to acquire the first processed partial signal, or to modulate the phase of the first spectral coefficient, to acquire the first processed partial signal, and a phase of the second spectral coefficient, to acquire a second processed partial signal; and

the combining step uses a synthesis filterbank adapted to combine the processed first partial signal and the second partial signal to acquire the output audio signal as a time domain signal, or to combine the processed first partial signal and the processed second partial signal to acquire the output audio signal as a time domain signal.

29. A non-transitory computer readable medium storing a computer program comprising program code for performing the method in accordance with claim 26 when the computer program runs on a computer.

27. A method for processing a multi channel audio signal, the method comprising:

receiving a first channel audio signal, a second channel audio signal, a third channel audio signal, a fourth channel audio signal and a fifth channel audio signal being part of a multi channel audio signal;

decorrelating the first channel audio signal and the second channel audio signal of the five channel audio signal using a method for decorrelating a first channel audio signal and a second channel audio signal, to acquire a first decorrelated output audio signal and a second decorrelated output audio signal, the method comprising;

deriving an output audio signal from an input audio signal using a method for deriving an output audio signal from an input audio signal, the method comprising;

extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

time-variantly modifying the first partial signal in dependence on a time-variant modulation function produced by a modulation function generator, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or time-variantly modifying only the first partial signal in dependence on the time-variant modulation function produced by the modulation function generator, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

combining the first processed partial signal and the second processed partial signal, or combining the first processed partial signal and the second partial signal, to acquire an output audio signal, wherein the first channel audio signal forms the input audio signal, and wherein the output audio signal forms the first decorrelated output audio signal; and

providing the second channel audio signal or a signal derived therefrom as the second decorrelated output audio signal;

wherein

the method for decorrelating is executed such that audio contents in the first audio frequency range of the first decorrelated output audio signal and of the second decorrelated output audio signal comprise a larger time-variant phase shift modulation amplitude with respect to each other than audio contents in the second audio frequency range of the first decorrelated output audio signal and of the second decorellated audio output signal, to acquire two signals of a decorrelated multi channel signal;

decorrelating the third channel audio signal and the fourth channel audio signal of the multi channel audio signal using a method for decorrelating a first channel audio signal and a second channel audio signal, to acquire a first decorrelated output audio signal and a second decorrelated output audio signal, the method comprising;

deriving an output audio signal from an input audio signal according to the method for deriving an output audio signal from an input audio signal, the method comprising;

extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

time-variantly modifying the first partial signal, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or time-variantly modifying only the first partial signal, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

combining the first processed partial signal and the second processed partial signal, or combining the first processed partial signal and the second partial signal, to acquire an output audio signal, wherein the first channel audio signal forms the input audio signal, and wherein the output audio signal forms the first decorrelated output audio signal; and

providing the second channel audio signal or a signal derived therefrom as the second decorrelated output audio signal;

the method for decorrelating is executed such that audio contents in the first audio frequency range of the first decorrelated output audio signal and of the second decorrelated output audio signal comprise a larger time-variant phase shift modulation amplitude with respect to each other than audio contents in the second audio frequency range of the first decorrelated output audio signal and of the second decorellated audio output signal, to acquire two further signals of the processed multi channel signal; and

decorrelating the fifth channel audio signal of the multi channel audio signal using a method for deriving an output audio signal from an input audio signal, the method comprising;

extracting a first partial signal and a second partial signal from the input audio signal, the first partial signal being descriptive of an audio content in a first audio frequency range, the second partial signal being descriptive of an audio content in a second audio frequency range, the first audio frequency range comprising higher frequencies compared to the second audio frequency range;

time-variantly modifying the first partial signal, to acquire a first processed partial signal, and the second partial signal, to acquire a second processed partial signal, so that a modulation amplitude of a time variant phase shift or time variant delay applied to the first partial signal is higher than a modulation amplitude of a time variant phase shift or a time variant delay applied to the second partial signal, or time-variantly modifying only the first partial signal, so that a time variant phase shift or a time variant delay is applied to the first partial signal; and

combining the first processed partial signal and the second processed partial signal, or combining the first processed partial signal and the second partial signal, to acquire an output audio signal, to acquire a fifth signal of the processed multi channel signal;

the extracting steps use an analysis filterbank adapted to decompose the input audio signal into at least two spectral coefficients, a first spectral coefficient being descriptive of the audio content in the first audio frequency range, the first spectral coefficient forming the first partial signal, a second spectral coefficient being descriptive of the audio content in the second audio frequency range, the second spectral coefficient forming the second partial signal;

the time-variantly modifying steps include modulating a phase at least of the first spectral coefficient, to acquire the first processed partial signal, or to modulate the phase of the first spectral coefficient, to acquire the first processed partial signal, and a phase of the second spectral coefficient, to acquire a second processed partial signal; and

the combining steps use a synthesis filterbank adapted to combine the processed first partial signal and the second partial signal to acquire the output audio signal as a time domain signal, or to combine the processed first partial signal and the processed second partial signal to acquire the output audio signal as a time domain signal.

30. A non-transitory computer readable medium storing a computer program comprising program code for performing the method in accordance with claim 27 when the computer program runs on a computer.