System and method for providing single microphone noise suppression fallback

• US 20090220107A1
• Filed: 02/29/2008
• Published: 09/03/2009
• Est. Priority Date: 02/29/2008
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Audience Corporation
• Inventors: Murgia, Carlo, Solbach, Ludger, Avendano, Carlos, Every, Mark
• Granted Patent: US 8,194,882 B2
• Time in Patent Office: Days
• US Class Current: 381/94.700
• CPC Class Codes: G10L 2021/02165 Two microphones, one receiving mainly the noise signal and the other one mainly the speech signal
• CPC Class Codes: G10L 21/0208 Noise filtering

Abstract

Systems and methods for providing single microphone noise suppression fallback are provided. In exemplary embodiments, primary and secondary acoustic signals are received. A single microphone noise estimate may be generated based on the primary acoustic signal, while a dual microphone noise estimate may be generated based on the primary and secondary acoustic signals. A combined noise estimate based on the single and dual microphone noise estimates is then determined. Using the combined noise estimate, a gain mask may be generated and applied to the primary acoustic signal to generate a noise suppressed signal. Subsequently, the noise suppressed signal may be output.

Claims

1. A method for providing single microphone noise suppression fallback, comprising:

receiving primary and secondary acoustic signals;

generating a single microphone noise estimate based on the primary acoustic signal;

generating a dual microphone noise estimate based on the primary and secondary acoustic signals;

determining a combined noise estimate based on the single and dual microphone noise estimates;

generating a gain mask based on the combined noise estimate applying the gain mask to the primary acoustic signal to generate a noise suppressed signal; and

outputting the noise suppressed signal.

2. The method of claim 1 wherein generating the single noise estimate comprises utilizing minimum statistics tracking.

3. The method of claim 1 wherein determining the combined noise estimate comprises determining a stationary noise estimate.

4. The method of claim 1 wherein determining the combined noise estimate comprises determining a non-stationary noise estimate.

5. The method of claim 1 wherein determining the combined noise estimate comprises selecting a maximum value between stationary and non-stationary noise estimates.

6. The method of claim 1 further comprising determining an inter-microphone level difference between the primary acoustic signal and a secondary acoustic signal.

7. The method of claim 6 wherein generating the dual microphone noise estimate comprises utilizing the inter-microphone level difference.

8. The method of claim 6 further comprising smoothing and mapping the inter-microphone level difference.

9. The method of claim 1 further comprising utilizing a secondary microphone likelihood indicator to determine a weighting factor to apply to the single microphone noise estimate.

10. A system providing one-microphone noise suppression fallback, comprising:

acoustic sensors configured to receive a primary and a secondary acoustic signal;

a single microphone noise estimate module configured to generate a single microphone noise estimate based on the primary acoustic signals;

a dual microphone noise estimate module configured to generate a dual microphone noise estimate based on the primary and secondary acoustic signals;

a noise estimate integrator configured to determine a combined noise estimate based on the single and dual microphone noise estimates;

a filter module configured to generate a gain mask based on the combined noise estimate; and

a masking module configured to apply the gain mask to the primary acoustic signal to generate a noise suppressed signal.

11. The system of claim 10 further comprising an inter-microphone level difference module configured to generate an inter-microphone level difference based on the primary and secondary acoustic signals.

12. The system of claim 10 wherein the noise estimate integrator comprises a weighting module configured to determine a weighting factor to apply to the single microphone noise estimate.

13. The system of claim 10 wherein the noise estimate integrator further comprises a stationary noise estimate module configured to determine a stationary noise estimate based on the single microphone noise estimate.

14. The system of claim 10 wherein the noise estimate integrator further comprises a non-stationary noise estimate module configured to determine a non-stationary noise estimate based on the dual microphone noise estimate.

15. The system of claim 10 wherein the noise estimate integrator comprises a maximizer module configured to determine the combined noise estimate based on a maximum value between stationary microphone noise estimates.

16. The system of claim 10 further comprising a microphone likelihood module configured to detected reliability of the secondary microphone.

17. The system of claim 10 wherein the single microphone noise estimate module comprises a minimum statistics tracker.

18. A machine readable medium having embodied thereon a program, the program providing instructions for a method for providing one-microphone noise suppression fallback, the method comprising:

receiving primary and secondary acoustic signals;

generating a single microphone noise estimate based on the primary acoustic signal;

generating a dual microphone noise estimate based on the primary and secondary acoustic signals;

determining a combined noise estimate based on the single and dual microphone noise estimates;

generating a gain mask based on the combined noise estimate applying the gain mask to the primary acoustic signal to generate a noise suppressed signal; and

outputting the noise suppressed signal.

19. The machine readable medium of claim 18 wherein determining the combined noise estimate comprises determining a stationary noise estimate and non-stationary noise estimate.

20. The machine readable medium of claim 19 wherein determining the combined noise estimate comprises selecting a maximum value between the stationary and non-stationary noise estimates.

21. A machine readable medium having embodied thereon a program, for providing instructions for a method for single microphone noise suppression fallback, the method comprising:

receiving primary and secondary acoustic signals;

generating a single microphone speech or noise estimate based on either the primary or the secondary acoustic signal;

generating a dual microphone speech or noise estimate based on both the primary acoustic signal and the secondary acoustic signal;

determining a combined speech estimate or a combined noise estimate based on the single and dual microphone estimates;

filtering either the primary or secondary acoustic signal using the combined speech estimate or the combined noise estimate to obtain a noise suppressed signal; and

outputting the noise suppressed signal.