Satellite microphones for improved speaker detection and zoom

• US 9,071,895 B2
• Filed: 11/19/2012
• Issued: 06/30/2015
• Est. Priority Date: 08/12/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Cutler, Ross G.
• Primary Examiner(s): WOO, STELLA L
• Application Number:
  US13/681,075
• Publication Number:
  US 20130093831A1
• Time in Patent Office: 953 Days
• Field of Search: 348 1401- 1416
• US Class Current: 1/1
• CPC Class Codes: H04N 7/142 Constructional details of the terminal equipment, e.g. arrangements of the camera and the display
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
• CPC Class Codes: H04R 1/406 microphones
• CPC Class Codes: H04R 2420/07 Applications of wireless loudspeakers or wireless microphones
• 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 5/027 Spatial or constructional arrangements of microphones, e.g. in dummy heads

Abstract

Architecture for exploiting satellite microphones and employing other techniques of conference room camera/microphone systems to significantly improve the true positive rate (reduce false positives) in sound source localization (SSL). Techniques for realizing the improvement include using an LED emitter to determine the precise location of the satellite microphones on a table, using the base SSL and external sounds to determine the approximate location of the satellite microphone on the table, using the satellite microphone phase to improve the SSL performance, using the satellite microphone amplitude to improve the active speaker detector (ASD) performance, and using the satellite microphones to estimate camera zoom.

Claims

1. A method of detecting a speaker comprising:

extending audio range of a local microphone array of a conferencing base station using satellite microphones;

computing location information for location of the satellite microphones relative to the conferencing base station;

processing the location information and audio signals of the local microphone array and the satellite microphones for automatic detection of a speaker relative to the conferencing base station; and

estimating location distances of the satellite microphones from the conferencing base station based, at least in part, on panoramic camera images of the satellite microphones.

2. The method of claim 1, further comprising locating the satellite microphones on opposing regions of the conferencing base station according to corresponding nominal locations and estimating location of a satellite microphone relative to the regions based on audio power.

3. The method of claim 1, further comprising estimating location distances of the satellite microphones from the conferencing base station based on vertical field of view of a camera, and head height of the camera.

4. The method of claim 3, further comprising computing sound source localization (SSL) based in part on distance and elevation information for each of the satellite microphones relative to the conferencing base station.

5. The method of claim 1, further comprising processing amplitude of the audio signals of the satellite microphones and half-plane location of the satellite microphones for detection of the speaker.

6. The method of claim 1, further comprising automatically estimating and imposing a zoom setting for a video camera of the conferencing base station based on amplitude of the audio signals of the satellite microphones and local microphone array.

7. The method of claim 1, further comprising automatically estimating and imposing a binary zoom setting for a video camera of the conferencing base station based on amplitude of the audio signals of the local microphone array.

8. The method of claim 1, further comprising determining whether to exercise zoom on a video camera of the conferencing base station based on estimated distance of a satellite microphone for the conferencing base station and audio signal time delay between the satellite microphone and the local microphone array.

9. A system for detecting a speaker including:

a conferencing base station including a local microphone array;

a plurality of satellite microphones connected to the conferencing base station; and

a computing device configured to

compute location information for location of the satellite microphones relative to the conferencing base station,

detect a speaker relative to the conferencing base station based on the location information and audio signals received from the local microphone array and the plurality of satellite microphones, and

estimate location distances of each of the plurality of satellite microphones from the conferencing base station based, at least in part, on panoramic camera images of each of the plurality of satellite microphones.

10. The system for detecting a speaker according to claim 9, where the computing device being further configured to

locate the plurality of satellite microphones on opposing regions of the conferencing base station according to corresponding nominal locations and

estimate the location of a satellite microphone relative to the opposing regions based on audio power.

11. The system for detecting a speaker according to claim 9, where the computing device being further configured to estimate location distances of each of the plurality of satellite microphones from the conferencing base station based on vertical field of view of a camera, and head height of the camera.

12. The system for detecting a speaker according to claim 11, where the computing device being further configured to compute sound source localization (SSL) based in part on a distance and an elevation for each of the plurality of satellite microphones relative to the conferencing base station.

13. The system for detecting a speaker according to claim 9, where the computing device being further configured to process an amplitude of the audio signals from the plurality of satellite microphones and a half-plane location of the plurality of satellite microphones for detection of the speaker.

14. The system for detecting a speaker according to claim 9, where the computing device being further configured to estimate and impose a zoom setting for a video camera of the conferencing base station based on an amplitude of the audio signals of the plurality of satellite microphones and the local microphone array.

15. The system for detecting a speaker according to claim 9, where the computing device being further configured to estimate and impose a binary zoom setting for a video camera of the conferencing base station based on an amplitude of the audio signals of the local microphone array.

16. The system for detecting a speaker according to claim 9, where the computing device being further configured to determine whether to exercise zoom on a video camera of the conferencing base station based on an estimated distance of a satellite microphone of the plurality of satellite microphones of the conferencing base station and an audio signal time delay between the satellite microphone of the plurality of satellite microphones and the local microphone array.

17. One of an optical disk, a magnetic storage device or a non-volatile memory device having computer-executable instructions stored thereon that, when executed by a computer, cause the computer to:

estimate the location distances of each of a plurality of satellite microphones from a conferencing base station is based, at least in part, on panoramic camera images of each of a plurality of satellite microphones; and

process the location distances and audio signals of a local microphone array on the conferencing base station and the plurality of satellite microphones for detection of a speaker relative to the conferencing base station.

18. The optical disk, magnetic storage device or non-volatile memory device according to claim 17, further comprising computer-executable instructions stored thereon that, when executed by the computer, further cause the computer to at least one of:

estimate the location distances of each of the plurality of satellite microphones from the conferencing base station based on vertical field of view of a camera, and head height of the camera;

or

compute sound source localization (SSL) based in part on a distance and an elevation for each of the plurality of satellite microphones relative to the conferencing base station.