Granted Patent B2

US 12,249,326 · App. 17/483,190 · Granted Mar 11, 2025

Inventors: John Usher (Beer, GB); Steven Goldstein (Delray Beach, FL); Marc Boillot (Plantation, FL)

Assignee: ST Case1Tech, LLC

G10L15/22G10L25/78H04R3/04G10L2015/223

Abstract

Claims (54)

1. A device, comprising:

a first ambient microphone configured to generate a first acoustic signal;

a second ambient microphone configured to generate a second acoustic signal;

a speaker configured to play an audio content signal;

a memory that stores instruction; and

a processor operatively connected to the first ambient microphone, the second ambient microphone, and the speaker, wherein the first and second ambient microphones and the speaker and the memory and the processor are part of a single device, wherein the processor executes the instructions to perform operations, the operations comprising:

generating a first spectral characteristics of the first acoustic signal;

generating a second spectral characteristics of the second acoustic signal;

selecting a first portion of the first spectral characteristics;

selecting a second portion of the second spectral characteristics;

detecting a user's voice based on an analysis of the first acoustic signal and the second acoustic signal wherein the analysis is performed by comparing the first portion to the second portion;

generating a modified audio signal by reducing the volume of the audio content signal if the user's voice has been detected;

mixing the first acoustic signal or a modified first acoustic signal, with the audio content signal or the modified audio signal, to generate a mixed audio content signal; and

sending the mixed audio content signal to the speaker.

2. The device according to claim 1 , wherein the operation further comprises:

detecting a background noise level by analyzing at least one of the first acoustic signal or the second acoustic signal or a combination of both.

3. The device of claim 1 , wherein the modified first acoustic signal is generated by applying a gain value to the first acoustic signal.

4. The device of claim 1 , wherein the modified first acoustic signal is generated by applying a filter to the first acoustic signal.

5. The device of claim 1 , wherein the modified first acoustic signal is generated by applying a filter or a gain to the first acoustic signal to generate an adjusted first acoustic signal, then generating the modified first acoustic signal by applying a gain to the adjusted first acoustic signal if a filter was applied to the first acoustic signal to generate the adjusted first acoustic signal, or generating the modified first acoustic signal by applying a filter to the adjusted first acoustic signal if a gain was applied to the first acoustic signal to generate the adjusted first acoustic signal.

6. The device of claim 1 , wherein the operations further comprise:

receiving a user's command, wherein the user's command is at least one of a voice command, a non-voice audio command, a visual command, or a combination thereof.

7. The device of claim 6 , wherein the operations further comprise:

calculating a cross correlation between at the first and second acoustic signals.

8. The device of claim 7 , wherein the operations further comprise:

extracting a peak value of the cross-correlation and a timing of the peak occurs.

9. The device of claim 8 , wherein the operations further comprise:

determining if the timing of the peak is within a predetermined time range.

10. The device of claim 9 , wherein the operations further comprise:

generating a message that the user voice is detected if the timing of the peak is within the predetermined time range.

11. The device of claim 1 , wherein the operation of detecting the voice includes an analysis of whether a sound pressure level of the first acoustic signal or the second acoustic signal or a combination of both is above a threshold and if above the threshold the level-comparison is performed to detect whether the user is speaking.

12. The device of claim 1 , wherein the operations further comprise:

analyzing the voice to detect a voice command if the user's voice has been detected.

13. The device of claim 1 , wherein the operations further comprise:

generating a filtered sound signal by filtering at least one of the first acoustic signal, the second acoustic signal, or a combination thereof, and

presenting a notification based on the filtered sound signal.

14. The device of claim 1 , wherein operations further comprise:

increasing a gain of at least one of the first acoustic signal, the second acoustic signal, or a combination thereof.

15. The device of claim 1 , wherein operations further comprise:

obtaining a speaking level associated with the user's voice by using a spectral characteristic of the voice if the user's voice has been detected.

16. The device according to claim 1 , wherein the gain is less than 1.

17. The device according to claim 16 , wherein the modified audio signal has a lower pressure level when emitted from a speaker than when the audio signal is emitted from the same speaker.

18. The device according to claim 1 , wherein the analysis uses at least one of a coherence analysis or correlation analysis or a cross-correlation or a combination thereof.

19. The device according to claim 1 , wherein the analysis uses at least one of level-comparison or spectral difference or a combination thereof.

20. A method comprising: receiving a first acoustic signal from a first microphone;

receiving a second acoustic signal from a second microphone;

generating first spectral characteristics of the first acoustic signal;

generating second spectral characteristics of the second acoustic signal;

selecting a first portion of the first spectral characteristics;

selecting a second portion of the second spectral characteristics;

comparing the first portion to the second portion to detect if a user's is speaking;

generating a modified audio content signal by reducing the volume of an audio content signal if the user is speaking;

generating a modified first acoustic signal by filtering the first acoustic signal if the user is speaking;

generating a mixed acoustic signal by combining the modified first acoustic signal with the modified audio content signal; and

sending the mixed acoustic signal to the speaker while the user is speaking or a neighboring talker is speaking to the user.

Assignments (9)

Continuity (5)

References Cited (400)

Cited By (1)

US 12,574,682