Multi-channel microphone signal gain equalization based on evaluation of cross talk components
View Patent ↗Gain mismatch and related problems can be solved by a system and method that applies an automatic microphone signal gain equalization without any direct absolute reference or calibration phase. The system and method performs the steps of receiving, by a computing device, a speech signal from a speaking person via a plurality of microphones, determining a speech signal component in the time-frequency domain for each microphone of the plurality of microphones, calculating an instantaneous cross-talk coupling matrix based on the speech signal components across the microphones, estimating gain factors based on calculated cross-talk couplings and a given expected cross-talk attenuation, limiting the gain factors to appropriate maximum and minimum values, and applying the gain factors to the speech signal used in the control path to control further speech enhancement algorithms or used in the signal path for direct influence on the speech enhanced audio output signal.
1. A computer-implemented method comprising:
receiving, by a computing device, at least one signal from each microphone of a plurality of microphones, wherein each microphone is associated with a respective spatial zone, and wherein the at least one signal comprises a speech signal component from a speaker in the respective spatial zone;
determining the speech signal component in the time-frequency domain for the at least one signal from each microphone;
calculating instantaneous cross-talk couplings based on each speech signal component;
estimating gain factors based on the calculated cross-talk couplings and a given expected cross-talk attenuation to yield estimated gain factors;
limiting the estimated gain factors to a range between a maximum value and a minimum value to yield final gain factors; and
applying the final gain factors to the the at least one signal from each microphone.
2. The method of claim 1 , wherein the final gain factors are applied to a control path.
3. The method of claim 1 , wherein the final gain factors are applied to a signal path.
4. The method of claim 1 , further comprising: detecting a spatial zone corresponding to an active speaker, wherein the spatial zone is detected using a trigger.
5. The method of claim 4 , wherein the trigger is at least one trigger selected from the group consisting of: Coherence-to-Diffuse-Ratio that indicates proximity of the microphone to the speaker, relative time delays between microphones, a Signal-to-Noise-Ratio smoothed over time, zone activity indicated by voice biometrics, and visual information provided by a camera.
6. The method of claim 1 , wherein each microphone is an array of microphones associated with a different spatial zone.
7. The method of claim 1 , wherein the expected cross-talk attenuation is defined by at least one existent acoustic environment factor selected from the group consisting of: microphone position, speaker position, dimension, reverberation, and combinations.
8. A computer program product residing on a non-transitory computer readable storage medium having a plurality of instructions stored thereon which, when executed across one or more processors, causes at least a portion of the one or more processors to perform operations comprising:
receiving, by a computing device, at least one signal from each microphone of a plurality of microphones, wherein each microphone is associated with a respective spatial zone, and wherein the at least one signal comprises a speech signal component from a speaker in the respective spatial zone;
determining the speech signal component in the time-frequency domain for the at least one signal from each microphone;
calculating instantaneous cross-talk couplings based on each speech signal component;
estimating gain factors based on the calculated cross-talk couplings and a given expected cross-talk attenuation to yield estimated gain factors;
limiting the estimated gain factors to a range between a maximum value and a minimum value to yield final gain factors; and
applying the final gain factors to the the at least one signal from each microphone.
9. The computer program product of claim 8 , wherein the final gain factors are applied to a control path.
10. The computer program product of claim 8 , wherein the final gain factors are applied to a signal path.
11. The computer program product of claim 8 , further comprising: detecting a spatial zone corresponding to an active speaker, wherein the spatial zone is detected using a trigger.
12. The computer program product of claim 11 , wherein the trigger is at least one trigger selected from the group consisting of: Coherence-to-Diffuse-Ratio that indicates proximity of the microphone to the speaker, relative time delays between microphones, a Signal-to-Noise-Ratio smoothed over time, zone activity indicated by voice biometrics, and visual information provided by a camera.
13. The computer program product of claim 8 , wherein each microphone is an array of microphones associated with a different spatial zone.
14. The computer program product of claim 8 , wherein the expected cross-talk attenuation is defined by at least one existent acoustic environment factor selected from the group consisting of: microphone position, speaker position, dimension, reverberation, and combinations.
15. A computing system including one or more processors and one or more non-transitory memories storing program instructions that, when executed, perform operations comprising:
receiving, by a computing device, at least one signal from each microphone of a plurality of microphones, wherein each microphone is associated with a respective spatial zone, and wherein the at least one signal comprises a speech signal component from a speaker in the respective spatial zone;
determining the speech signal component in the time-frequency domain for each at least one signal;
calculating instantaneous cross-talk couplings based on each speech signal component;
estimating gain factors based on the calculated cross-talk couplings and a given expected cross-talk attenuation to yield estimated gain factors;
limiting the estimated gain factors to a range between a maximum value and a minimum value to yield final gain factors; and
applying the final gain factors to the the at least one signal from each microphone.
16. The computing system of claim 15 , wherein the final gain factors are applied to a control path.
17. The computing system of claim 15 , wherein the final gain factors are applied to a signal path.
18. The computing system of claim 15 , further comprising: detecting a spatial zone corresponding to an active speaker, wherein the spatial zone is detected using a trigger.
19. The computing system of claim 18 , wherein the trigger is at least one trigger selected from the group consisting of: Coherence-to-Diffuse-Ratio that indicates proximity of the microphone to the speaker, relative time delays between microphones, a Signal-to-Noise-Ratio smoothed over time, zone activity indicated by voice biometrics, and visual information provided by a camera.
20. The computing system of claim 15 , wherein each microphone is an array of microphones associated with a different spatial zone.
21. The computing system of claim 15 , wherein the expected cross-talk attenuation is defined by at least one existent acoustic environment factor selected from the group consisting of: microphone position, speaker position, dimension, reverberation, and combinations.