Unified microphone pre-processing system and method

• US 9,171,551 B2
• Filed: 12/22/2011
• Issued: 10/27/2015
• Est. Priority Date: 01/14/2011
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations LLC (General Motors Company)
• Inventors: Tzirkel-Hancock, Eli, Tsimhoni, Omer
• Primary Examiner(s): YEN, ERIC L
• Application Number:
  US13/334,920
• Publication Number:
  US 20120185247A1
• Time in Patent Office: 1,405 Days
• US Class Current: 1/1
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2021/02166 Microphone arrays; Beamforming
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: H04M 9/08 Two-way loud-speaking telephone systems with means for conditioning the signal, e.g. for suppressing echoes for one or both directions of traffic
• CPC Class Codes: H04R 1/406 microphones
• 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 unified microphone pre-processing system includes a plurality of microphones arranged within a vehicle passenger compartment, a processing circuit or system configured to receive signals from one or more of the plurality of microphones, and the processing circuit configured to enhance the received signals for use by at least two of a telephony processing application, an automatic speech recognition processing application, and a noise cancellation processing application. The method includes receiving signals from one or more of a plurality of microphones arranged within a vehicle passenger compartment, and enhancing the received signals for use by at least two of a telephony processing application, an automatic speech recognition processing application, and a noise cancellation processing application. A computer readable medium containing executable instructions to cause a processor to perform a method in accordance with an embodiment of the invention is also described.

Claims

1. A processor implemented method for using a single set of spatially disparate microphones for plurality of audio applications, the method comprising:

receiving signals from the microphones;

processing the received signals to produce one or more processed-signals;

providing the processed-signals to the plurality of audio applications, the plurality of audio applications comprising at least;

an active noise cancellation application, and one of;

a telephony application and an automatic speech recognition application; and

outputting inverted noise signals via one or more speakers, wherein the inverted noise signals are generated by the active noise cancellation application;

further wherein the processing comprises;

performing enhancement to the received signals, the enhancement configured for acoustic echo cancellation;

performing noise reduction and residual echo reduction, both by;

steering at least one of multiple reception beams to a speaker'"'"'s mouth,

using signal enhancement technique, and

using signal reconstruction technique;

performing a speech recognition enhancement pre-processing on the received signals, wherein the speech recognition enhancement pre-processing is configured to optimize a metric of at least one of;

Mel cepstrum distance or word recognition rate; and

performing sound pressure measurements near a head of one or more occupants.

2. The method of claim 1, wherein the processing further comprises at least one of:

adaptively beamforming two or more reception beams of the microphones, responsive to changes in acoustic conditions; and

directing at least one of the adaptively beamformed reception beams to at least one of the occupants.

3. A unified microphone pre-processing system comprising:

a plurality of microphones arranged within a vehicle passenger-compartment;

one or more loudspeakers arranged within the vehicle passenger-compartment;

a memory unit comprising;

an active noise cancellation processing application (ANCPA) configured to output inverted noise signals via at least one of the loudspeakers, and

at least one of;

telephony processing application and automatic speech recognition processing application; and

a processor configured for;

receiving signals from the plurality of microphones,

performing enhancement to the received signals, the enhancement configured for acoustic echo cancellation,

performing noise reduction and residual echo reduction, both by;

steering at least one of multiple reception beams to a speaker'"'"'s mouth,

using signal enhancement technique, and

using signal reconstruction technique,

performing a speech recognition enhancement pre-processing on the received signals, wherein the speech recognition enhancement pre-processing is configured to optimize a metric of at least one of;

Mel cepstrum distance or word recognition rate, and

performing sound pressure measurements near a head of one or more occupants;

generating one or more enhanced-signals; and

outputting the inverted noise signals via the one or more speakers;

wherein;

at least one of the enhanced-signals is configured to serve as input to the ANCPA, and

at least one of the enhanced-signals is configured to serve as input to at least one of;

the telephony processing application and the automatic speech recognition processing application.

4. The system of claim 3, wherein the processing-system comprises a pre-processing unit comprising:

an analog-to-digital converter configured to convert the received-signals to digital-signals;

a memory unit configured for storing; and

a control processor configured to perform pre-processing enhancement techniques on the digital-signals.

5. The system of claim 3, wherein the processing-system comprises a beamformer unit configured for adaptive construction of two or more reception beams of the plurality of microphones, responsive to changes in acoustic conditions of the vehicle passenger:

compartment.

6. The system of claim 4, wherein the processing-system further comprising at least one of:

an acoustic echo cancellation pre-processing unit, configured to reduce acoustic echo; and

a speech recognition pre-processing unit configured to enhance the digital-signals and to optimize a metric of at least one of;

Mel cepstrum distance and word recognition rate.

7. The system of claim 3, wherein the plurality of microphones are arranged in a predetermined geometrical pattern.

8. The system of claim 7, wherein the geometrical pattern is semispherical.

9. The system of claim 3, wherein the processing-system further comprises a noise cancellation pre-processing unit configured for:

locating a target occupant;

retrieving microphone array parameters for locations near the target occupant; and

interpolating the array parameters in order to generate the at least one of the enhanced-signals configured to serve as input to the ANCPA.

10. The system of claim 9, further comprising a camera configured for the locating of the target occupant.

11. The system of claim 9, wherein at least one of the microphones is an omni-directional microphone.

12. A processor implemented method for unified microphone pre-processing, the method comprising:

receiving signals from a plurality of microphones, wherein the microphones are arranged within a vehicle passenger-compartment;

pre-processing the received signals, wherein the pre-processing comprises;

performing enhancement to the received signals, the enhancement configured for acoustic echo cancellation,

performing noise reduction and residual echo reduction, both by;

steering at least one of multiple reception beams to a speaker'"'"'s mouth,

using signal enhancement technique, and

using signal reconstruction technique,

performing a speech recognition enhancement pre-processing on the received signals, wherein the speech recognition enhancement pre-processing is configured to optimize a metric of at least one of;

Mel cepstrum distance or word recognition rate, and

performing sound pressure measurements near a head of one or more occupants;

generating one or more enhanced-signals, wherein;

at least one of the enhanced-signals is configured for serving as input to an active noise cancellation processing application (ANCPA) stored in a memory, and

at least one of the enhanced-signals is configured for serving as input to at least one of;

telephony processing application stored in a memory and automatic speech recognition processing application stored in a memory; and

outputting inverted noise signals via one or more loudspeakers arranged within the vehicle passenger-compartment, wherein the inverted noise signals are generated by the ANCPA.

13. The method of claim 12, further comprising converting the received-signals from analog-signals to digital-signals, and wherein the pre-processing further comprises performing enhancement techniques on the digital-signals.

14. The method of claim 12, further comprising at least one of:

adaptively beamforming two or more reception beams of the microphones, responsive to changes in acoustic condition of the vehicle passenger-compartment; and

steering at least one of the adaptively beamformed reception beams of the microphones to a driver and one or more passengers in a vehicle passenger-compartment.

15. The method of claim 12, wherein the plurality of microphones are arranged in a predetermined geometrical pattern.

16. The method of claim 12, further comprising:

locating a target occupant;

retrieving microphone array parameters for locations near the target occupant; and

interpolating the array parameters in order to generate the at least one of the enhanced-signals configured for serving as input to the ANCPA.

17. The method of claim 12, wherein at least one of the microphones is an omni-directional microphone.