Multichannel acoustic echo reduction

• US 8,385,557 B2
• Filed: 06/19/2008
• Issued: 02/26/2013
• Est. Priority Date: 06/19/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Tashev, Ivan Jelev, Acero, Alejandro, Madhu, Nilesh
• Primary Examiner(s): Paul, Disler
• Application Number:
  US12/141,941
• Publication Number:
  US 20090316923A1
• Time in Patent Office: 1,713 Days
• Field of Search: 381/66, 381/83, 381/93, 381/971.11, 37940601-40609, 379/406.14
• US Class Current: 381/66
• CPC Class Codes: H04M 9/082 using echo cancellers echo cancellers per se H04B3/23
• CPC Class Codes: H04R 3/005 for combining the signals of two or more microphones specially adapted for hearing aids H04R25/407
• CPC Class Codes: H04R 3/02 for preventing acoustic reaction , i.e. acoustic oscillatory feedback specially adapted for hearing aids H04R25/453

Abstract

A multichannel acoustic echo reduction system is described herein. The system includes an acoustic echo canceller (AEC) component having a fixed filter for each respective combination of loudspeaker and microphone signals and having an adaptive filter for each microphone signal. For each microphone signal, the AEC component modifies the microphone signal to reduce contributions from the outputs of the loudspeakers based at least in part on the respective adaptive filter associated with the microphone signal and the set of fixed filters associated with the respective microphone signal.

Claims

1. A multichannel acoustic echo reduction system comprising:

a signal receiving component that receives a plurality of loudspeaker signals that drive a plurality of respective loudspeakers, and further receives one or more microphone signals generated by one or more respective microphones based at least in part on outputs from the loudspeakers;

an acoustic echo canceller (AEC) component having a fixed filter for each respective combination of loudspeaker and microphone signals and having an adaptive filter for each microphone signal, wherein for each microphone signal, the AEC component modifies the microphone signal to reduce contributions from the outputs of the loudspeakers based at least in part on the respective adaptive filter associated with the microphone signal and the set of fixed filters associated with the respective microphone signal; and

a calibration component that provides coefficients that configure each of the fixed filters, wherein the calibration component sequentially includes a calibration signal in each loudspeaker signal and determines the coefficients for each fixed filter based at least in part on the microphone signals received during time periods for which the calibration signals are included in the loudspeaker signals.

2. The system of claim 1, wherein when the receiving component receives L loudspeaker signals and M microphone signals, the AEC component generates a quantity of L×

M fixed filters and a quantity of M adaptive filters.

3. The system of claim 1, wherein the signal receiving component receives a plurality of microphone signals generated by a plurality of respective microphones, wherein the plurality of microphone signals is greater than the plurality of loudspeaker signals, further comprising a tracking component that determines relative movement between at least one loudspeaker and microphone and triggers the calibration component to provide updated coefficients for the fixed filters based at least in part upon the determination.

4. The system of claim 1, wherein each fixed filter is associated with a different combination of one microphone signal and one loudspeaker signal, wherein each fixed filter produces an output based at least in part on the respective associated loudspeaker signal and the coefficients configured for the respective fixed filter.

5. The system of claim 4, wherein the signal receiving component receives a plurality of microphone signals generated by a plurality of microphones, wherein each adaptive filter is associated with a respective microphone signal, wherein for each microphone signal, the respective associated adaptive filter produces an output based at least in part on a combination of the outputs from the fixed filters associated with the respective microphone signal and respective configuration values updated based at least in part on the respective microphone signal and respective previous configuration values associated with the respective adaptive filter.

6. The system of claim 5, wherein the adaptive filters produce respective outputs based at least in part on at least one AEC algorithm selected from the group comprising least mean square LMS, normalized least mean square (NMLS), or recursive least squares (RLS).

7. The system of claim 5, wherein for each microphone signal, the respective associated adaptive filter includes a finite impulse response (FIR) filter with a predetermined length to produce an output corresponding to an approximation of a transfer function between the respective microphone signal and the combination of the outputs from the fixed filters associated with the respective microphone signal.

8. The system of claim 5, wherein for each microphone signal, the AEC component combines the output from the respective associated adaptive filter with the respective microphone signal to produce a respective filtered microphone signal.

9. The system of claim 8, further comprising a beamformer component that produces a filtered output based at least in part on the filtered microphone signals produced by the AEC component.

10. The system of claim 9, further comprising an acoustic echo suppression (AES) component that produces a filtered output based at least in part on the filtered output produced by the beamformer component.

11. The system of claim 10, wherein the AES component produces outputs based at least in part on at least one AES algorithm selected from the group comprising a Wiener gain algorithm, an Ephraim, or a Malah algorithm.

12. A method comprising:

receiving a plurality of loudspeaker signals that drive a plurality of respective loudspeakers;

sequentially including a calibration signal in each loudspeaker signal;

receiving a one or more microphone signals generated by one or more respective microphones based at least in part on outputs from the loudspeakers;

with an acoustic echo canceller (AEC) component having a fixed filter for each respective combination of loudspeaker and microphone signals and an adaptive filter for each microphone signal, modifying each microphone signal to reduce contributions from the outputs of the loudspeakers based at least in part on the respective adaptive filter associated with each microphone signal and the set of fixed filters associated with the respective microphone signal; and

determining coefficients that configure each fixed filter based at least in part on the microphone signals received during time periods for which the calibration signals are included in the loudspeaker signals.

13. The method of claim 12, further comprising generating a quantity of L×

M fixed filters and a quantity of M adaptive filters when receiving L loudspeaker signals and M microphone signals.

14. The method of claim 12, wherein each fixed filter is associated with a different combination of one microphone signal and one loudspeaker signal, wherein each adaptive filter is associated with a respective microphone signal, further comprising:

receiving a plurality of microphone signals generated by a plurality of respective microphones;

producing with each fixed filter, an output based at least in part on the respective associated loudspeaker signal and the coefficients configured for the respective fixed filter;

producing for each microphone signal with the respective associated adaptive filter, an output based at least in part on a combination of the outputs from the fixed filters associated with the respective microphone signal and respective configuration values updated based at least in part on the respective microphone signal and respective previous configuration values associated with the respective adaptive filter; and

combining for each microphone signal, the output from the respective associated adaptive filter with the respective microphone signal to produce a respective filtered microphone signal.

15. The method of claim 14, further comprising producing a first filtered output based at least in part on the filtered microphone signals produced by the adaptive filters.

16. The method of claim 15, further comprising producing a second filtered output based at least in part on the first filtered output.

17. A computer-readable data storage device comprising instructions that, when executed by a processor, cause the processor to perform acts comprising:

receiving a quantity of L loudspeaker signals that drive a quantity of L respective loudspeakers;

receiving a quantity of M microphone signals generated by a quantity of M respective microphones based at least in part on outputs from the loudspeakers;

for an acoustic echo canceller (AEC) generating a quantity of L×

M fixed filters and a quantity of M adaptive filters, wherein each fixed filter is associated with a different combination of one microphone signal and one loudspeaker signal, wherein each adaptive filter is associated with a respective microphone signal;

sequentially including a calibration signal in each loudspeaker signal;

determining coefficients that configure each fixed filter based at least in part on the microphone signals received during time periods for which the calibration signals are included in the loudspeaker signals;

producing with each fixed filter, an output based at least in part on the respective associated loudspeaker signal and the coefficients configured for the respective fixed filter;

producing for each microphone signal with the respective associated adaptive filter, an output based at least in part on a combination of the outputs from the fixed filters associated with the respective microphone signal and respective configuration values updated based at least in part on the respective microphone signal and respective previous configuration values associated with the respective adaptive filter;

combining for each microphone signal, the output from the respective associated adaptive filter with the respective microphone signal to produce a respective filtered microphone signal in which contributions from the outputs of the loudspeakers are reduced;

producing with a beamformer component, a filtered output based at least in part on the filtered microphone signals produced by the adaptive filters; and

producing with an acoustic echo suppression (AES) component, a filtered output based at least in part on the filtered output produced by the beamformer component.