System and method for performing automatic gain control using an accelerometer in a headset

Patent metadata

Field	Value
Publication number	US-9997173-B2
Application number	US-201615069808-A
Country	US
Kind code	B2
Filing date	Mar 14, 2016
Priority date	Mar 14, 2016
Publication date	Jun 12, 2018
Grant date	Jun 12, 2018

Abstract

A method performing automatic gain control (AGC) using an accelerometer in a headset starts with an accelerometer-based voice activity detector (VADa) generating a VADa output based on (i) acoustic signals received from at least one microphone included in a pair of earbuds and (ii) data output by at least one accelerometer that is included in the pair of earbuds. The at least one accelerometer detects vibration of the user's vocal chords. The headset includes the pair of earbuds. An AGC controller then performs automatic gain control (AGC) on the acoustic signals from the at least one microphone based on the VADa output. Other embodiments are also described.

First claim

The invention claimed is:

1. A method of performing automatic gain control (AGC) using an accelerometer in a headset comprising:
generating by an accelerometer-based voice activity detector (VADa) a VADa output based on (i) acoustic signals received from at least one microphone included in a pair of earbuds and (ii) data output by at least one accelerometer that is included in the pair of earbuds, the at least one accelerometer to detect vibration of a user's vocal chords, the headset including the pair of earbuds,
wherein generating the VADa output comprises at least one of:
(i) computing a power envelope of at least one of x, y, z direction signals generated by the at least one accelerometer, and setting the VADa output to indicate that a user's voiced speech is detected if the power envelope is greater than a power envelope threshold and setting the VADa output to indicate that the user's voiced speech is not detected if the power envelope is less than the power envelope threshold, and/or
(ii) computing a normalized cross-correlation between any pair of the x, y, z direction signals generated by the at least one accelerometer, setting the VADa output to indicate that the user's voiced speech is detected if the normalized cross-correlation is greater than a threshold within a predetermined delay, and setting the VADa output to indicate that the user's voiced speech is not detected if the normalized cross-correlation is less than the threshold; and

performing automatic gain control (AGC) by an AGC controller on the acoustic signals from the at least one microphone based on the VADa output.

2. The method of claim 1 , wherein generating the VADa output further comprises:
applying an exponential decay function and a smoothing function to the VADa output.

3. The method of claim 1 , wherein the VADa is included in an equalizer.

4. The method of claim 3 , further comprising:
generating by an equalizer an equalized accelerometer signal based on (i) the acoustic signals received from the at least one microphone, and (ii) the data output by the at least one accelerometer.

5. The method of claim 4 , further comprising:
generating by a beamformer a voice beam signal and a noise beam signal based on the acoustic signals received from the at least one microphone included in the pair of earbuds;
generating by a noise suppressor a clean beamformer signal by suppressing noise from the voice beam signal, wherein the noise suppressor receives the voice beam signal and the noise beam signal.

6. The method of claim 5 , further comprising:
generating by an omni selector an omni signal based on the acoustic signals received from the at least two microphones included in the pair of earbuds.

7. The method of claim 6 , further comprising:
performing spectral mixing of (i) the equalized accelerometer signal, (ii) the clean beamformer signal, and (iii) the omni signal to generate a mixed signal.

8. The method of claim 7 , wherein performing AGC by the AGC controller includes:
performing AGC by the AGC controller on the mixed signal based on the VADa output.

9. The method of claim 8 , wherein performing AGC by the AGC controller includes:
controlling a fast energy tracker and a slow energy tracker by the VADa output, to generate a fast gain and a slow gain.

10. The method of claim 9 , further comprising:
performing encoding by a speech codec on the AGC output based on the VADa output.

11. The method of claim 10 , further comprising:
generating a pitch estimate by a pitch detector based on (i) the acoustic signals received from at least one microphone included in a pair of earbuds and (ii) the data output by at least one accelerometer that is included in the pair of earbuds.

12. The method of claim 11 , wherein the pitch detector generates the pitch estimate based on an autocorrelation method and using the output from the at least one accelerometer, wherein the pitch estimate is obtained by (i) using an X, Y, or Z signal generated by the at least one accelerometer that has a highest power level or (ii) using a combination of the X, Y, and Z signals generated by the at least one accelerometer.

13. The method of claim 12 , wherein performing AGC by the AGC controller includes:
performing AGC by the AGC controller on the mixed signal based on the VADa output and the pitch estimate.

14. A system for performing automatic gain control (AGC) using an accelerometer in a headset comprising:
the headset including a pair of earbuds, wherein each of the earbuds includes at least one microphone and at least one accelerometer to detect vibration of a user's vocal chords,
an accelerometer-based voice activity detector (VADa) to generate a VADa output based on (i) acoustic signals received from the at least one microphone included in the pair of earbuds and (ii) data output by the at least one accelerometer that is included in the pair of earbuds,
wherein generating the VADa output comprises at least one of:
(i) computing a power envelope of at least one of x, y, z direction signals generated by the at least one accelerometer, and setting the VADa output to indicate that a user's voiced speech is detected if the power envelope is greater than a power envelope threshold and setting the VADa output to indicate that the user's voiced speech is not detected if the power envelope is less than the power envelope threshold, and/or
(ii) computing a normalized cross-correlation between any pair of the x, y, z direction signals generated by the at least one accelerometer, setting the VADa output to indicate that the user's voiced speech is detected if the normalized cross-correlation is greater than a threshold within a delay, and setting the VADa output to indicate that the user's voiced speech is not detected if the normalized cross-correlation is less than the threshold; and

an AGC controller to perform AGC on the acoustic signals from the at least one microphone based on the VADa output.

15. The system of claim 14 , wherein the VADa generates the VADa output by:
applying an exponential decay function and a smoothing function to the VADa output.

16. The system of claim 14 , further comprising an equalizer to generate an equalized accelerometer signal based on (i) the acoustic signals received from the at least one microphone, and (ii) the data output by the at least one accelerometer.

17. The system of claim 16 , further comprising:
a beamformer to generate a voice beam signal and a noise beam signal based on the acoustic signals received from the at least one microphone included in the pair of earbuds;
a noise suppressor to generate a clean beamformer signal by suppressing noise from the voice beam signal, wherein the noise suppressor receives the voice beam signal and the noise beam signal.

18. The system of claim 17 , further comprising:
a omni selector generating an omni signal based on the acoustic signals received from the at least two microphones included in the pair of earbuds.

19. The system of claim 18 , further comprising:
a spectral mixer to perform spectral mixing of (i) the equalized accelerometer signal, (ii) the clean beamformer signal, and (iii) the omni signal to generate a mixed signal.

20. The system of claim 19 , wherein the AGC controller performs AGC on the mixed signal based on the VADa output.

21. The system of claim 20 , wherein the AGC controller performs AGC by controlling a fast energy tracker and a slow energy tracker by the VADa output to generate a fast gain and a slow ga

Assignees

• Apple Inc

Inventors

• Dusan Sorin V
• Paquier Baptiste
• Lindahl Aram
• Biruski Dubravko

Classifications

• G10L21/0388

  Details of processing therefor · CPC title

• G10L21/034

  Automatic adjustment · CPC title

• G10L2025/783

  based on threshold decision · CPC title

• G10L25/90

  Pitch determination of speech signals · CPC title

• G10L2021/02166

  Microphone arrays; Beamforming · CPC title