SPATIAL ADAPTATION IN MULTI-MICROPHONE SOUND CAPTURE

• US 20130315403A1
• Filed: 02/10/2012
• Published: 11/28/2013
• Est. Priority Date: 02/10/2011
• Status: Active Grant

• Current Assignee: Dolby International AB (Dolby Laboratories Incorporated)
• Original Assignee: Dolby International AB (Dolby Laboratories Incorporated)
• Inventors: Samuelsson, Leif Jonas
• Granted Patent: US 9,538,286 B2
• Time in Patent Office: Days
• US Class Current: 381/56
• CPC Class Codes: G10L 21/0232 Processing in the frequency domain
• CPC Class Codes: H04R 2430/20 Processing of the output signals of the acoustic transducers of an array for obtaining a desired directivity characteristic H04R2203/12 takes precedence
• CPC Class Codes: H04R 29/006 Microphone matching
• 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 3/04 for correcting frequency response

Abstract

A spatial adaptation system for multiple-microphone sound capture systems and methods thereof are described. A spatial adaptation system includes an inference and weight module configured to receive a inputs. The inputs based on two or more input signals captured by at least two microphones. The inference and weight module to determine one or more weight values base on at least one of the inputs. The spatial adaptation system also including a noise magnitude ratio update module coupled with the inference and weight module. The noise magnitude ratio update module to determine an updated noise target based on the one or more weight values from the inference and weight module.

Claims

1. A spatial adaptation system comprising:

an inference and weight module configured to receive a plurality of inputs based on two or more input signals captured by at least two microphones, said inference and weight module to determine one or more weight values based on at least one of said plurality of inputs; and

a noise magnitude ratio update module coupled with said inference and weight module, said noise magnitude ratio update module to determine an updated noise target based on said one or more weight values from said inference and weight module.

2. The system of claim 1 further comprising a spatial feature module coupled with said inference and weight module, said spatial feature module to determine one or more spatial features based on said two or more input signals.

3. The system of claim 2, wherein said inference and weight module determines one or more weight values based on said one or more spatial features determined by said spatial feature module.

4. The system of claim 3, wherein said one or more spatial features include magnitude ratio based on said two or more input signals, phase difference based on said two or more input signals, and coherence based on said two or more input signals.

5. The system of claim 1, wherein at least two of said plurality of inputs are represented as frequency-domain signals.

6. The system of claim 5, wherein said inference and weight module uses one or more Gaussian mixture models to classify said at least two of said plurality of inputs.

7. The system of claim 6, wherein said inference and weight module classifies said at least two of said plurality of inputs into one or more classes including clean signal, noise, and interferer.

8. The system of claim 5, wherein inference and weight module determines a maxima follower of magnitude ratios based on said at least two of said plurality of inputs.

9. The system of claim 8, wherein said inference and weight module discriminates between near-field source and far-field noise based on said maxima follower.

10. A system for spatial adaptation comprising:

a plurality of microphones to capture a sound source;

an input signal conversion module coupled with at least two of said plurality of microphones to convert said sound source captured into at least two frequency-domain signals;

a spatial feature module coupled with input conversion module, said spatial feature module to determine one or more spatial features based on said at least two frequency-domain signals;

an inference and weight module coupled with said spatial feature module, said inference and weight module to determine one or more inferences based on said one or more spatial features;

a spatial adaptation module coupled with said input signal conversion module, said spatial feature module, and said inference and weight module, said spatial adaptation module to determine a noise target based on said one or more spatial features; and

a matching multiplier coupled with said input signal conversion module, said spatial feature module, and said spatial adaptation module, said matching multiplier to determine a matched signal based on said noise target and at least one of said at least two frequency-domain signals.

11. The system of claim 10 further comprising a beamformer module coupled with said input signal conversion module, said spatial feature module, said beamformer to determine a combined signal based on said matched signal and at least one of said at least two frequency-domain signals.

12. The system of claim 11 further comprising a combined signal multiplier and an inference and weight module coupled with said combined signal multiplier, said spatial feature module, and said spatial adaptation module, said spatial adaptation module to determine one or more spatial features, said inference and weight module to determine a gain based on one or more spatial features and said desired source, and said combined signal multiplier to determine an output signal based on said gain and said combined signal.

13. The system of claim 10, wherein said spatial adaptation module determines said noise target only if said one or more inferences indicate a voice frame.

14. A method for spatial adaptation, the method comprising:

Receiving two or more frequency-domain signals based on signals captured by at least two microphones;

Determining one or more spatial features based on each of said two or more frequency-domain signals;

Determining one or more weight values based on said plurality of spatial features; and

Updating a noise target based on said one or more determined weight values.

15. The method of claim 14 further comprising determining an interferer based on one or more of said plurality of spatial features.

16. The method of claim 15, wherein said spatial features include magnitude ratio based on said two or more frequency-domain signals, phase difference based on said two or more frequency-domain signals, and coherence based on said two or more frequency-domain signals.

17. The method of claim 15 further comprising classifying said two or more frequency-domain signals using at least one Gaussian mixture model.

18. A memory device readable by a machine, embodying a program of instructions executable by the machine to perform a method for suppressing noise in one or more of at least first and second channels, the method comprising:

Receiving two or more signals based on two or more input signals captured by at least two microphones;

Determining a plurality of spatial features based on said two or more signals;

Determining one or more weight values based on said plurality of spatial features; and

Updating a noise target based on said one or more determined weight values.

19. The method of claim 18 further comprising determining an interferer based on one or more of said plurality of spatial features.

20. The method of claim 18, wherein said spatial features include magnitude ratio of based on said two or more frequency-domain signals, phase difference based on said two or more frequency-domain signals, and coherence based on said two or more frequency-domain signals.

21. The method of claim 18 further comprising classifying said two or more frequency-domain signals using at least one Gaussian mixture model.

22. Method for discrimination between near-field and far-field sound sources based on minima following of magnitude ratios.

23. As claim 22 with minima following based on minima search over buffers containing magnitude ratios.

24. As claim 22 with minima following based on adding a bias term to the minima estimate.

27. As claim 22 with maxima following based on subtracting a bias term to the maxima estimate.

25. Method for discrimination between near-field and far-field sound sources based on maxima following of magnitude ratios.

26. As claim 25 with maxima following based on minima search over buffers containing magnitude ratios.

28. Method for discrimination between near-field and far-field sound sources based on minima and maxima following of magnitude ratios.

29. The use of the postSNR feature inThe inference andThe update weight computation

30. The mechanism to adapt noise targets to clean speech.

31. The use of double minima following when doing (single channel) noise estimation in claims dependent on claims covering the postSNR feature which is based on the noise estimator.

32. A MR maxima follower can be used to discriminate between near-field and far-field sources.