Noisy environment communication enhancement system

• US 9,002,028 B2
• Filed: 04/16/2014
• Issued: 04/07/2015
• Est. Priority Date: 05/09/2003
• Status: Active Grant

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Haulick, Tim, Schmidt, Gerhard Uwe
• Primary Examiner(s): Cao, Phat X
• Application Number:
  US14/254,007
• Publication Number:
  US 20140233757A1
• Time in Patent Office: 356 Days
• Field of Search: 381/86, 381/91, 381/92, 381/122
• US Class Current: 381/92
• CPC Class Codes: G10L 2021/02082 the noise being echo, reverberation of the speech
• 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 2021/02166 Microphone arrays; Beamforming
• CPC Class Codes: H04M 9/082 using echo cancellers echo cancellers per se H04B3/23
• CPC Class Codes: H04R 2410/05 Noise reduction with a separate noise microphone
• CPC Class Codes: H04R 2430/23 Direction finding using a sum-delay beam-former
• CPC Class Codes: H04R 2430/25 Array processing for suppression of unwanted side-lobes in directivity characteristics, e.g. a blocking matrix
• CPC Class Codes: H04R 2499/13 Acoustic transducers and sound field adaptation in vehicles
• CPC Class Codes: H04R 3/005 for combining the signals of two or more microphones specially adapted for hearing aids H04R25/407

Abstract

A communication system enhances communications in a noisy environment. Multiple input arrays convert a voiced or unvoiced signal into an analog signal. A converter receives the analog signal and generates digital signals. A digital signal processor determines temporal and spatial information from the digital signals. The processed signals are then converted to audible sound.

Claims

1. A method, comprising:

receiving input signals from at least two microphone arrays each comprising at least two microphones to generate microphone signals;

supplying the microphone signals to a beamformer comprising a wanted signal beamformer and a blocking beamformer,

in the wanted signal beamformer, combining the microphone signals with a computer processor such that signals from a wanted direction corresponding to a first one of the microphone arrays and a given speaker are transmitted substantially without attenuation and signal components of directions other than the wanted direction are substantially attenuated; and

in the blocking beamformer, substantially suppressing at least one of the microphone signals from the wanted signal direction, and transmitting with minimum distortion at least one of the microphone signals from directions other than the wanted direction.

2. The method according to claim 1, further including identifying potential feedback frequencies and placing adaptive notch filters at the potential feedback frequencies.

3. The method according to claim 1, further including attenuating one of the microphones associated with non-active speakers for mitigating double-talk effects.

4. The method according to claim 1, further including comparing powers of signal outputs from the wanted signal beamformer and the blocking beamformer to determine a given one of a plurality of speech scenarios.

5. The method according to claim 4, wherein the plurality of speech scenarios include no speakers, one speaker, and one speaker located not in the wanted direction.

6. The method according to claim 1, further including enhancing the directivity of the wanted signal beamformer with an adaptive beamformer.

7. The method according to claim 6, wherein the adaptive beamformer comprises a general sidelobe canceller.

8. The method according to claim 6, further including adapting the adaptive beamformer only when no signal from the wanted signal direction is transmitted.

9. The method according to claim 1, further including detecting whether there are one or more speakers are active and outputting signals for the active speakers to loudspeakers.

10. The method according to claim 1, further including performing automatic gain control outputs of the beamformer based on distance between speakers and microphones.

11. The method according to claim 1, determining an output volume for loudspeakers of signals from the beamformer based upon background noise level.

12. The method according to claim 1, further comprising detecting speech activity for each microphone array, wherein detecting speech activity for a microphone array comprises:

determining a wanted signal power, a blocking signal power, and a background noise signal power, and

comparing the wanted signal power with the blocking signal power and the background noise signal power.

13. The method according to claim 1, further comprising selecting at least one output channel out of at least two output channels on which the processed signals are to be output.

14. A system, comprising:

a beamformer to receive input signals from at least two microphone arrays each comprising at least two microphones to generate microphone signals,

wherein the beamformer comprises;

a wanted signal beamformer to combine the microphone signals such that signals from a wanted direction corresponding to a first one of the microphone arrays and a given speaker are transmitted substantially without attenuation and signal components of directions other than the wanted direction are substantially attenuated; and

a blocking beamformer to substantially suppress at least one of the microphone signals from the wanted signal direction, and to transmit with minimum distortion at least one of the microphone signals from directions other than the wanted direction.

15. An article, comprising:

a non-transitory computer readable medium having stored instructions that enable a machine to;

receive input signals from at least two microphone arrays each comprising at least two microphones to generate microphone signals;

supplying the microphone signals to a wanted signal beamformer and a blocking beamformer,

in the wanted signal beamformer, combining the microphone signals such that signals from a wanted direction corresponding to a first one of the microphone arrays and a given speaker are transmitted substantially without attenuation and signal components of directions other than the wanted direction are substantially attenuated; and

in the blocking beamformer, substantially suppressing at least one of the microphone signals from the wanted signal direction, and transmitting with minimum distortion at least one of the microphone signals from directions other than the wanted direction.

16. The article according to claim 15, further including instructions for identifying potential feedback frequencies and placing adaptive notch filters at the potential feedback frequencies.

17. The article according to claim 15, further including instructions for attenuating one of the microphones associated with non-active speakers for mitigating double-talk effects.

18. The article according to claim 15, further including instructions for comparing powers of signal outputs from the wanted signal beamformer and the blocking beamformer to determine a given one of a plurality of speech scenarios.

19. The article according to claim 18, wherein the plurality of speech scenarios include no speakers, one speaker, and one speaker located not in the wanted direction.

20. The article according to claim 15, further including instructions for enhancing the directivity of the wanted signal beamformer with an adaptive beamformer.