Processing audio signals

1. A method of processing a multi-channel audio signal, the method comprising:

determining a location of sound sources within the audio signal;

applying a rotation operation to the audio signal, a direction of the rotation operation dependent on the location of the sound sources in the signal;

generating a rotated audio signal;

wherein determining the location of sound sources within the signal comprises comparing the signal to pre-determined calibration information associated with a recording device; and

transforming the audio signal into the frequency domain;

wherein generating the rotated audio signal comprises transforming the signal into the time domain after the rotation operation has been applied to the signal.

2. The method of claim 1 ,

wherein determining the location of sound sources comprises classifying sound sources as located in a first region or a second region relative to a recording device.

3. The method of claim 2 ,

wherein the first region is in front of the recording device and the second region is behind the recording device.

4. The method of claim 2

wherein applying the rotation operation to the signal comprises:

separating the signal into a first signal and a second signal, the first signal comprising the sound sources classified as located in the first region, and the second signal comprising sound sources classified as located in the second region;

applying a first rotation operation to the first signal to generate a rotated first signal; and

applying a second rotation operation to the second signal to generate a rotated second signal.

5. The method of claim 4 ,

wherein generating the rotated audio signal comprises

combining the rotated first signal with the rotated second signal to form the rotated audio signal.

6. The method of claim 4 :

wherein the first and second rotation operations rotate signals in opposite directions by a same rotation angle.

7. The method of claim 1 ,

wherein determining the location of sound sources comprises

determining the location of each of a plurality of sound sources within the audio signal.

8. The method of claim 7 ,

wherein determining the location of sound sources further comprises

determining a direction of arrival of each sound source at the recording device.

9. The method of claim 7 ,

wherein applying a rotation operation comprises:

generating a rotation matrix relating the sound source locations to desired sound source locations; and

applying the rotation matrix to the signal.

10. The method of claim 1 ,

wherein determining a location of sound sources within the signal comprises

estimating the location of the sound sources using a regression algorithm.

11. The method of claim 1 :

wherein determining the location of sound sources within the signal comprises

establishing the location of sound sources by determining the values of s for which the function ∥y−Hs∥ 2 +λ∥s∥ 2 is a minimum,

where y is a matrix of size M×1 representing the observed signals, with M the number of channels in the audio signal, s is a vector with size N×1 representing a vector of the sound source locations with N the number of possible sound source candidates, H is the calibration matrix with size M×N, and λ is the Lagrange multiplier.

12. An audio processing device comprising:

an input for receiving a multi-channel audio signal;

a processor operable to perform the method of claim 1 on the signal; and

an output for outputting the rotated audio signal.

13. An article of manufacture including at least one non-transitory, tangible machine readable storage medium containing executable machine instructions for causing a processor to perform the method according to claim 1 .

14. A method of processing a multi-channel audio signal, the method comprising:

determining a location of sound sources within the audio signal;

applying a rotation operation to the audio signal, a direction of the rotation operation dependent on the location of the sound sources in the signal; and

generating a rotated audio signal;

wherein determining the location of sound sources comprises classifying sound sources as located in a first region or a second region relative to a recording device.

wherein applying a rotation operation to the signal comprises:

separating the signal into a first signal and a second signal, the first signal comprising the sound sources classified as located in the first region, and the second signal comprising sound sources classified as located in the second region;

applying a first rotation operation to the first signal to generate a rotated first signal; and

applying a second rotation operation to the second signal to generate a rotated second signal.

15. An audio processing device comprising:

an input for receiving a multi-channel audio signal;

a processor operable to,

determine a location of sound sources within the audio signal;

apply a rotation operation to the audio signal, a direction of the rotation operation dependent on the location of the sound sources in the signal; and

generate a rotated audio signal;

wherein determine the location of sound sources includes,

classify sound sources as located in a first region or a second region relative to a recording device.

wherein apply the rotation operation to the signal includes,

separate the signal into a first signal and a second signal, the first signal including the sound sources classified as located in the first region, and the second signal including sound sources classified as located in the second region;

apply a first rotation operation to the first signal to generate a rotated first signal; and

apply a second rotation operation to the second signal to generate a rotated second signal; and

an output for outputting the rotated audio signal.

16. The device of claim 15 ,

wherein generating the rotated audio signal comprises

combining the rotated first signal with the rotated second signal to form the rotated audio signal.

17. The device of claim 15 :

wherein the first and second rotation operations rotate signals in opposite directions by a same rotation angle.

18. The device of claim 15 ,

wherein determining the location of sound sources comprises

determining the location of each of a plurality of sound sources within the audio signal.