Acoustic event detection

Patent metadata

Field	Value
Publication number	US-12308045-B2
Application number	US-202318243804-A
Country	US
Kind code	B2
Filing date	Sep 8, 2023
Priority date	Dec 10, 2021
Publication date	May 20, 2025
Grant date	May 20, 2025

Abstract

A system may include a first acoustic event detection (AED) component configured to detect a predetermined set of acoustic events, and include a second AED component configured to detect custom acoustic events that a user configures a device to detect. The first and second AED components are configured to perform task-specific processing, and may receive as input the same acoustic feature data corresponding to audio data that potentially represents occurrence of one or more events. Based on processing by the first and second AED components, a device may output data indicating that one or more acoustic events occurred, where the acoustic events may be a predetermined acoustic event and/or a custom acoustic event.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
receiving first audio data representing occurrence of a first acoustic event;
determining first encoded data representing the first acoustic event;
receiving second audio data representing audio detected by a device;
determining, by processing the second audio data and the first encoded data using a first acoustic event detection (AED) component configured to detect occurrence of one or more acoustic events from a first set of acoustic events, first event detection data representing a first likelihood that at least one acoustic event from the first set of acoustic events is represented in the second audio data;
determining, by processing the first audio data using a second AED component configured to detect occurrence of one or more acoustic events from a second set of acoustic events, second event detection data representing a second likelihood that at least one acoustic event from the second set of acoustic events is represented in the second audio data;
determining, based at least in part on the first event detection data and the second event detection data, that at least one of the first acoustic event from the first set of acoustic events or a second acoustic event from the second set of acoustic events is represented in the second audio data; and
determining output data indicating that at least one of the first acoustic event or the second acoustic event occurred.

2. The computer-implemented method of claim 1 , further comprising:
determining, using the second audio data, acoustic feature data,
wherein processing the second audio data using the first AED component comprises processing the first acoustic feature data using the first AED component, and
wherein processing the second audio data using the second AED component comprises processing the acoustic feature data using the second AED component.

3. The computer-implemented method of claim 1 , wherein the first AED component comprises a machine learning component.

4. The computer-implemented method of claim 1 , wherein the second AED component comprises a machine learning component.

5. The computer-implemented method of claim 1 , wherein the device is associated with a profile and wherein the method further comprises:
storing the first encoded data in a manner associated with the profile.

6. The computer-implemented method of claim 1 , wherein processing by the second AED component comprises processing the first encoded data with respect to second encoded data corresponding to the second audio data.

7. A system comprising:
at least one processor; and
at least one memory comprising instructions that, when executed by the at least one processor, cause the system to:
receive first audio data representing occurrence of a first acoustic event;
determine first encoded data representing the first acoustic event;
receive second audio data representing audio detected by a device;
determine, by processing the second audio data and the first encoded data using a first acoustic event detection (AED) component configured to detect occurrence of one or more acoustic events from a first set of acoustic events, first event detection data representing a first likelihood that at least one acoustic event from the first set of acoustic events is represented in the second audio data;
determine, by processing the second audio data using a second AED component configured to detect occurrence of one or more acoustic events from a second set of acoustic events, second event detection data representing a second likelihood that at least one acoustic event from the second set of acoustic events is represented in the second audio data;
determine, based at least in part on the first event detection data and the second event detection data, that at least one of the first acoustic event from the first set of acoustic events or a second acoustic event from the second set of acoustic events is represented in the second audio data; and
determine output data indicating that at least one of the first acoustic event or the second acoustic event occurred.

8. The system of claim 7 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
determine, using the second audio data, acoustic feature data,
wherein the instructions that cause the system to process the second audio data using the first AED component comprise instructions that, when executed by the at least one processor, cause the system to process the acoustic feature data using the first AED component, and
wherein the instructions that cause the system to process the second audio data using the second AED component comprise instructions that, when executed by the at least one processor, cause the system to process the first acoustic feature data using the second AED component.

9. The system of claim 7 , wherein the first AED component comprises a machine learning component.

10. The system of claim 7 , wherein the second AED component comprises a machine learning component.

11. The system of claim 7 , wherein the device is associated with a profile and wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
store the first encoded data in a manner associated with the profile.

12. The system of claim 7 , wherein the instructions that cause the system to process by the second AED component comprise instructions that, when executed by the at least one processor, cause the system to process, by the second AED component, the first encoded data with respect to second encoded data corresponding to the second audio data.

13. The system of claim 7 , wherein:
processing by the first AED component is performed by the device; and
processing by the second AED component is performed by the device.

14. The computer-implemented method of claim 1 , wherein the first set of acoustic events include a custom set of events associated with the device.

15. The computer-implemented method of claim 1 , wherein the second set of acoustic events include a predetermined set of events.

16. The computer-implemented method of claim 1 , wherein the first AED component is a comparison-based AED component.

17. The computer-implemented method of claim 1 , wherein the second AED component is a classifier-based AED component.

18. The computer-implemented method of claim 1 , wherein the first AED component is configured by a user.

19. The system of claim 7 , wherein the first set of acoustic events include a custom set of events associated with the device.

20. The system of claim 7 , wherein the second set of acoustic events include a predetermined set of events.

Assignees

• Amazon Tech Inc

Inventors

• Tang Qingming
• Kao Chieh-Chi
• Zhang Qin
• Sun Ming
• Wang Chao
• Garg Sumit
• Chen Rong
• Droppo James Garnet
• Chang Chia-Jung

Classifications

• G10L2015/223

  Execution procedure of a spoken command · CPC title

• G10L2015/088

  Word spotting · CPC title

• G10L25/30

  using neural networks · CPC title

• G10L25/21

  the extracted parameters being power information · CPC title

• G10L15/22

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title