SPEECH RECOGNITION METHODS, DEVICES, AND SYSTEMS

• US 20150364135A1
• Filed: 06/11/2014
• Published: 12/17/2015
• Est. Priority Date: 06/11/2014
• Status: Active Grant

• Current Assignee: Ademco Inc. (Resideo Technologies, Inc.)
• Original Assignee: Honeywell International Inc.
• Inventors: Kolavennu, Soumitri N., Katuri, SrinivasaRao, Kulkarni, Amit
• Granted Patent: US 10,510,343 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/32 Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems
• CPC Class Codes: G10L 2021/02087 the noise being separate speech, e.g. cocktail party
• CPC Class Codes: G10L 2021/02166 Microphone arrays; Beamforming
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: H04R 3/005 for combining the signals of two or more microphones specially adapted for hearing aids H04R25/407

Abstract

Speech recognition methods, devices, and systems are described herein. One system includes a number of microphones configured to capture sound in an area, a digital signal processor configured to segregate the captured sound into a plurality of signals, wherein each respective signal corresponds to a different portion of the area, and an automatic speech recognition engine configured to separately process each of the plurality of signals to recognize a speech command in the captured sound.

Claims

1. A system for speech recognition, comprising:

a number of microphones configured to capture sound in an area;

a digital signal processor configured to segregate the captured sound into a plurality of signals, wherein each respective signal corresponds to a different portion of the area; and

an automatic speech recognition engine configured to separately process each of the plurality of signals to recognize a speech command in the captured sound.

2. The system of claim 1, wherein the number of microphones, the digital signal processor, and the automatic speech recognition engine are included in a sound recognition device.

3. The system of claim 1, wherein the digital signal processor and the automatic speech recognition engine are included in a sound recognition device, wherein the sound recognition device is at least one of:

a thermostat;

a security system; and

an alarm clock.

4. The system of claim 1, wherein:

the number of microphones and the digital signal processor are included in a sound recognition device; and

the automatic speech recognition engine is communicatively coupled to the sound recognition device via a network.

5. The system of claim 1, wherein:

the number of microphones are included in a sound recognition device; and

the digital signal processor and the automatic speech recognition engine are communicatively coupled to the sound recognition device via a network.

6. The system of claim 1, wherein the area is a room.

7. A method for speech recognition, comprising:

capturing sound in an area, wherein the captured sound includes a speech command;

segregating the captured sound into a plurality of signals, wherein each respective signal corresponds to a different portion of the area; and

recognizing the speech command by separately processing each of the plurality of signals.

8. The method of claim 7, wherein:

the captured sound includes background noise in the area; and

recognizing the speech command includes separating the speech command from the background noise.

9. The method of claim 8, wherein the background noise includes speech noise and non-speech noise.

10. The method of claim 7, wherein each respective different portion of the area is a different angular segment of the area.

11. The method of claim 7, wherein the method includes:

capturing the sound using a sound recognition device;

segregating the captured sound into the plurality of signals using the sound recognition device; and

recognizing the speech command by separately processing each of the plurality of signals using the sound recognition device.

12. The method of claim 11, wherein the method includes adjusting, by the sound recognition device, operation of the sound recognition device or another device based on the recognized speech command.

13. The method of claim 7, wherein the method includes:

capturing the sound using a sound recognition device;

segregating the captured sound into the plurality of signals using the sound recognition device;

transmitting the plurality of signals from the sound recognition device to an automatic speech recognition engine via a network; and

recognizing the speech command by separately processing each of the plurality of signals using the automatic speech recognition engine.

14. The method of claim 13, wherein the method includes:

transmitting an instruction to adjust operation of the sound recognition device from the automatic speech recognition engine to the sound recognition device, wherein the instruction is based on the recognized speech command; and

adjusting, by the sound recognition device, the operation of the sound recognition device based on the instruction.

15. The method of claim 7, wherein the method includes:

capturing the sound using a sound recognition device;

transmitting the captured sound from the sound recognition device to a digital signal processor via a network;

segregating the captured sound into the plurality of signals using the digital signal processor;

transmitting the plurality of signals from the digital signal processor to an automatic speech recognition engine coupled to the digital signal processor; and

recognizing the speech command by separately processing each of the plurality of signals using the automatic speech recognition engine.

16. A system for speech recognition, comprising:

a number of microphones configured to capture sound in an area;

a digital signal processor configured to segregate the captured sound into a plurality of signals, wherein each respective signal corresponds to a different angular portion of the area with respect to the number of microphones; and

an automatic speech recognition engine configured to separately process each of the plurality of signals to recognize a speech command in the captured sound.

17. The system of claim 16, wherein each respective different angular portion of the area has a same angular size with respect to the number of microphones.

18. The system of claim 16, wherein each respective different angular portion of the area has an angular size of thirty degrees with respect to the number of microphones.

19. The system of claim 16, wherein each respective different angular portion of the area has an angular size of less than thirty degrees with respect to the number of microphones.

20. The system of claim 16, wherein the system includes a plurality of automatic speech recognition engines configured to process the plurality of signals in parallel.