Location estimation of active speaker

• US 10,219,098 B2
• Filed: 09/18/2017
• Issued: 02/26/2019
• Est. Priority Date: 03/03/2017
• Status: Active Grant

• Current Assignee: BAR ILAN University, GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations LLC (General Motors Company)
• Inventors: Tzirkel-Hancock, Eli, Tourbabin, Vladimir, Malka, Ilan, Gannot, Sharon
• Primary Examiner(s): Chin, Vivian
• Assistant Examiner(s): Fahnert, Friedrich W
• Application Number:
  US15/707,299
• Publication Number:
  US 20180255418A1
• Time in Patent Office: 526 Days
• Field of Search: 381 56, 381 58, 381 66, 381 714, 381 86, 381 92, 381 98, 381120, 381302, 381313, 381315, 704215, 704233, 704240, 700 94
• CPC Class Codes: H04R 1/406 microphones
• CPC Class Codes: H04R 2201/401 2D or 3D arrays of transducers
• 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 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
• CPC Class Codes: H04S 7/305 Electronic adaptation of stereophonic audio signals to reverberation of the listening space H04S7/301 takes precedence

Abstract

A system and method to perform an estimation of a location of an active speaker in real time includes designating a microphone of an array of microphones as a reference microphone. The method includes storing a relative transfer function (RTF) for each microphone of the array of microphones other than the reference microphone associated with each potential location among potential locations as a set of stored RTFs, and obtaining a voice sample of the active speaker and obtaining a speaker RTF for each microphone of the array of microphones other than the reference microphone. The method also includes performing an RTF projection of the speaker RTF for each microphone on the set of stored RTFs, and determining one of the potential locations as the location of the active speaker based on the performing the RTF projection.

Claims

1. A method of performing an estimation of a location of an active speaker in real time, the method comprising:

designating any one microphone of an array of microphones as a reference microphone;

storing a relative transfer function (RTF) for each microphone of the array of microphones other than the reference microphone associated with each potential location among potential locations as a set of stored RTFs;

obtaining a voice sample of the active speaker and obtaining a speaker RTF for each microphone of the array of microphones other than the reference microphone;

performing an RTF projection of the speaker RTF for each microphone on the set of stored RTFs; and

Determining, using a processor, one of the potential locations as the location of the active speaker based on the performing the RTF projection, wherein the obtaining the speaker RTF for each microphone of the array of microphones other than the reference microphone includes computing, for each of the potential locations, a ratio of an acoustic transfer function of the voice sample at the microphone to an acoustic transfer function of the voice sample at the reference microphone.

2. The method according to claim 1, wherein the obtaining the voice sample is performed in real time.

3. The method according to claim 1, further comprising sampling a sound from each of the potential locations to obtain the set of stored RTFs.

4. The method according to claim 1, further comprising obtaining the set of stored RTFs as the RTF for each microphone of the array of microphones other than the reference microphone based on computing, for each of the potential locations, a ratio of an acoustic transfer function from one potential location among the potential locations to the microphone to an acoustic transfer function from the one potential location among the potential locations to the reference microphone.

5. The method according to claim 1, wherein the performing the RTF projection includes calculating a cosine distance between each speaker RTF and each RTF of the set of stored RTFs.

6. The method according to claim 5, wherein the determining the location of the active speaker is based on the maximum of the cosine distances.

7. The method according to claim 1, wherein the storing the set of stored RTFs for the potential locations includes storing the set of stored RTFs for each seat in an automobile.

8. The method according to claim 7, wherein the storing the set of stored RTFs is part of a calibration process performed for the automobile.

9. The method according to claim 7, wherein the storing the set of stored RTFs is part of a calibration process performed for a calibration automobile of a same model as the automobile.

10. A system to estimate a location of an active speaker, the system comprising:

a memory device configured to store a relative transfer function (RTF) for each microphone of an array of microphones other than a reference microphone associated with each potential location among potential locations as a set of stored RTFs, wherein the reference microphone is any one of the array of microphones; and

a processor configured to obtain a voice sample of the active speaker and obtain a speaker RTF for each microphone of the array of microphones other than the reference microphone, perform an RTF projection of the speaker RTF for each microphone on the set of stored RTFs, and determine one of the potential locations as the location of the active speaker based on the RTF projection, wherein the processor obtains the speaker RTF for each microphone of the array of microphones other than the reference microphone based on computing, for each of the potential locations, a ratio of an acoustic transfer function of the voice sample at the microphone to an acoustic transfer function of the voice sample at the reference microphone.

11. The system according to claim 10, wherein the processor obtains the voice sample in real time.

12. The system according to claim 10, wherein the processor samples a sound from each of the potential locations to obtain the set of stored RTFs.

13. The system according to claim 10, wherein the processor obtains the set of stored RTFs as the RTF for each microphone of the array of microphones other than the reference microphone based on computing, for each of the potential locations, a ratio of an acoustic transfer function from one potential location among the potential locations to the microphone to an acoustic transfer function from the one potential location among the potential locations to the reference microphone.

14. The system according to claim 10, wherein the processor performs the RTF projection by calculating a cosine distance between each speaker RTF and each RTF of the set of stored RTFs.

15. The system according to claim 14, wherein the processor determines the location of the active speaker based on the maximum of the cosine distances.

16. The system according to claim 10, wherein the memory device stores the set of stored RTFs for each seat in an automobile.

17. The system according to claim 16, wherein the memory device stores the set of stored RTFs as part of a calibration process performed for the automobile.

18. The system according to claim 16, wherein the memory device stores the set of stored RTFs as part of a calibration process performed for a calibration automobile of a same model as the automobile.