Audio source localization system and method
A system and method for performing audio source localization that uses minimum variance distortionless response (MVDR) beams or other super-directive beams that are more suitable for reducing noise, interference and/or acoustic coupling than the delay-and-sum beams conventionally used in performing audio source localization based on a Steered Response Power (SRP) approach. Since super-directive beams such as MVDR beams do not provide good properties for performing conventional SRP-based audio source localization, an embodiment of the present invention utilizes a different metric than maximum response power in order to determine which of a plurality of beams provides the best response for audio source localization. This metric involves identifying which beam provides the smallest measure of distortion with respect to a reference power or reference response.
1. A method for estimating a direction of arrival of sound waves emanating from an audio source, comprising:
receiving a plurality of audio signals produced by an array of microphones;
processing the plurality of audio signals in a beamformer to produce a plurality of responses corresponding respectively to a plurality of beams having different look directions;
calculating a measure of distortion for each response in the plurality of responses;
determining which response in the plurality of responses provides a smallest measure of distortion; and
identifying the look direction of the beam corresponding to the response that provides the smallest measure of distortion as the estimated direction of arrival.
2. The method of claim 1 , wherein the plurality of responses corresponds respectively to a plurality of superdirective beams having different look directions.
3. The method of claim 2 , wherein the plurality of responses corresponds respectively to a plurality of Minimum Variance Distortionless Response (MVDR) beams having different look directions.
4. The method of claim 3 , wherein the plurality of MVDR beams comprises a plurality of fixed MVDR beams.
5. The method of claim 3 , wherein the plurality of MVDR beams comprises a plurality of adaptive MVDR beams.
6. The method of claim 1 , wherein calculating the measure of distortion for each response in the plurality of responses comprises:
calculating an absolute difference between a power of each response in the plurality of responses and a reference power.
7. The method of claim 6 , wherein the reference power comprises a power of a response of a designated microphone in the array of microphones.
8. The method of claim 6 , wherein the reference power comprises an average response power of two or more designated microphones in the array of microphones.
9. The method of claim 1 , wherein calculating the measure of distortion for each response in the plurality of responses comprises:
calculating a power of a difference between each response in the plurality of responses and a reference response.
10. The method of claim 1 , wherein calculating the measure of distortion for a response in the plurality of responses comprises:
(a) calculating a measure of distortion for the response at each of a plurality of frequencies;
(b) summing the measures of distortion calculated in step (a).
11. The method of claim 1 , wherein calculating the measure of distortion for a response in the plurality of responses comprises:
(a) calculating a measure of distortion for the response at each of a plurality of frequencies;
(b) multiplying each measure of distortion calculated in step (a) by a frequency-dependent weight to produce a plurality of frequency-weighted measures of distortion; and
(c) summing the frequency-weighted measures of distortion calculated in step (b).
12. The method of claim 1 , further comprising:
steering a beam produced by a steerable beamformer toward the estimated direction of arrival; and
processing the plurality of audio signals in the steerable beamformer to produce an output audio signal for acoustic transmission.
13. The method of claim 1 , further comprising:
using the response that provides the smallest measure of distortion to produce an output audio signal for acoustic transmission.
14. A system for estimating a direction of arrival of sound waves emanating from an audio source, comprising:
an array of microphones;
a beamformer that processes a plurality of audio signals produced by the array of microphones to produce a plurality of responses corresponding respectively to a plurality of beams having different look directions;
a distortion calculator that calculates a measure of distortion for each response in the plurality of responses; and
a direction of arrival selector that determines which response in the plurality of responses provides a smallest measure of distortion and identifies the look direction of the beam corresponding to the response that provides the smallest measure of distortion as the estimated direction of arrival.
15. The system of claim 14 , wherein the plurality of responses corresponds respectively to a plurality of superdirective beams having different look directions.
16. The system of claim 15 , wherein the plurality of responses corresponds respectively to a plurality of Minimum Variance Distortionless Response (MVDR) beams having different look directions.
17. The system of claim 16 , wherein the plurality of MVDR beams comprises a plurality of fixed MVDR beams.
18. The system of claim 16 , wherein the plurality of MVDR beams comprises a plurality of adaptive MVDR beams.
19. The system of claim 14 , wherein the distortion calculator calculates the measure of distortion for each response in the plurality of responses by calculating an absolute difference between a power of each response in the plurality of responses and a reference power.
20. The system of claim 19 , wherein the reference power comprises a power of a response of a designated microphone in the array of microphones.
21. The system of claim 19 , wherein the reference power comprises an average response power of two or more designated microphones in the array of microphones.
22. The system of claim 14 , wherein the distortion calculator calculates the measure of distortion for each response in the plurality of responses by calculating a power of a difference between each response in the plurality of responses and a reference response.
23. The system of claim 14 , wherein the distortion calculator calculates the measure of distortion for a response in the plurality of responses by:
(a) calculating a measure of distortion for the response at each of a plurality of frequencies;
(b) summing the measures of distortion calculated in step (a).
24. The system of claim 14 , wherein the distortion calculator calculates the measure of distortion for a response in the plurality of responses by:
(a) calculating a measure of distortion for the response at each of a plurality of frequencies;
(b) multiplying each measure of distortion calculated in step (a) by a frequency-dependent weight to produce a plurality of frequency-weighted measures of distortion; and
(c) summing the frequency-weighted measures of distortion calculated in step (b).
25. The system of claim 14 , further comprising:
a steerable beamformer that is configured to steer a beam toward the estimated direction of arrival and to process the plurality of audio signals to produce an output audio signal;
acoustic transmission logic configured to transmit the output audio signal.
26. The system of claim 14 , further comprising:
acoustic transmission logic configured to transmit an output audio signal produced using the response that provides the smallest measure of distortion.
27. A computer program product comprising a computer-readable medium having computer program logic recorded thereon for enabling a processor to estimate a direction of arrival of sound waves emanating from an audio source, the computer program logic comprising:
first means for enabling the processor to receive a plurality of audio signals produced by an array of microphones;
second means for enabling the processor to process the plurality of audio signals to produce a plurality of responses corresponding respectively to a plurality of beams having different look directions;
third means for enabling the processor to calculate a measure of distortion for each response in the plurality of responses;
fourth means for enabling the processor to determine which response in the plurality of responses provides a smallest measure of distortion; and
fifth means for enabling the processor to identify the look direction of the beam corresponding to the response that provides the smallest measure of distortion as the estimated direction of arrival.