SPEECH RECOGNITION

• US 20140142934A1
• Filed: 11/21/2012
• Published: 05/22/2014
• Est. Priority Date: 11/21/2012
• Status: Active Grant

• Current Assignee: Empire Technology Development LLC (Allied Inventors Management, LLC)
• Original Assignee: Empire Technology Development LLC (Allied Inventors Management, LLC)
• Inventors: Kim, Seungil
• Granted Patent: US 9,251,804 B2
• Time in Patent Office: Days
• US Class Current: 704/226
• 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 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: H04W 4/025 using location based information parameters

Abstract

Technologies are generally described for a speech recognition scheme. In some examples, a method performed under control of a speech recognition system may include receiving, from a first device, first data including a first signal captured by the first device, first location information of the first device, and first time information corresponding to the captured first signal; cancelling first noise from the captured first signal based at least in part on the first location information and the first time information, and estimating a first voice signal of a first user of the first device, wherein the first noise is associated with a second voice signal of a second user of a second device located adjacent to the first device; and translating the first voice signal into a first command for the first device.

Claims

1. A speech recognition system, comprising:

a receiver configured to receive first data from a first device of a first user and second data from a second device of a second user,

wherein the first data includes a first voice signal captured by the first device, and first location information of the first device, and

wherein the second data includes a second voice signal captured by the second device, and second location information of the second device;

a noise canceller configured to cancel first noise from the captured first voice signal based at least in part on the first location information and the second location information, wherein the first noise is associated with the second voice signal of the second user of the second device; and

a speech recognizer configured to translate the first voice signal into a first command for the first device.

2. The speech recognition system of claim 1, further comprising:

a location comparator configured to compare the first location information and the second location information to determine whether the first device and the second device are adjacent to each other.

3. The speech recognition system of claim 1, wherein the first location information includes at least one of a GPS (Global Positioning System) coordinate of the first device, Wi-Fi access point information of the first device, cell information of a cellular network accessed by the first device, a location identification of the first device, and information of other devices detected by the first device, and

wherein the second location information includes a GPS coordinate of the second device, Wi-Fi access point information of the second device, cell information of a cellular network accessed by the second device, a location identification of the second device, and information of other devices detected by the second device.

4. The speech recognition system of claim 1, wherein the first data further includes first time information corresponding to the captured first voice signal, and the second data further includes second time information corresponding to the captured second voice signal, and

wherein the noise canceller is further configured to cancel the first noise from the captured first voice signal further based on the first time information and the second time information.

5. The speech recognition system of claim 4, further comprising:

a time comparator configured to compare the first time information and the second time information to determine whether the capture of the first voice signal and the capture of the second voice signal overlap in time,

wherein the first time information includes a start time of the first voice signal and an end time of the first voice signal, and

wherein the second time information includes a start time of the second voice signal and an end time of the second voice signal.

6. The speech recognition system of claim 1, wherein the noise canceller is further configured to cancel second noise from the captured second voice signal, and further configured to estimate the second voice signal, wherein the second noise is associated with the first voice signal, and

wherein the speech recognizer is further configured to translate the second voice signal into a second command for the second device.

7. The speech recognition system of claim 1, wherein the noise canceller utilizes an adaptive filtering algorithm.

8. The speech recognition system of claim 1, further comprising:

a voice comparator configured to calculate a correlation coefficient between the captured first voice signal and the captured second voice signal.

9. The speech recognition system of claim 1, further comprising:

a transmitter configured to transmit the first command to the first device.

10. A method performed under control of a speech recognition system, comprising:

receiving, from a first device of a first user, first data including a first voice signal captured by the first device, first location information of the first device, and first time information corresponding to the captured first voice signal;

receiving, from a second device of a second user, second data including a second voice signal captured by the second device;

cancelling first noise from the captured first voice signal based at least in part on the first location information and the first time information, wherein the first noise is associated with the second voice signal of the second user of the second device located adjacent to the first device; and

translating the first voice signal into a first command for the first device.

11. The method of claim 10, wherein the first location information includes at least one of a GPS (Global Positioning System) coordinate of the first device, Wi-Fi access point information of the first device, cell information of a cellular network accessed by the first device, a location identification of the first device, and information of other devices detected by the first device.

12. The method of claim 10, wherein the first time information includes a start time of the first voice signal and an end time of the first voice signal.

13. The method of claim 10, further comprising:

transmitting the first command to the first device.

14. The method of claim 10, further comprising:

receiving, from the second device, second data including the second voice signal captured by the second device, second location information of the second device and second time information corresponding to the captured second voice signal;

determining whether the first device and the second device are adjacent to each other based at least in part on the first location information and the second location information; and

determining whether the capture of the first voice signal and the capture of the second voice signal overlap in time based at least in part on the first time information and the second time information, when determining that the first device and the second device are adjacent to each other,

wherein the cancelling includes cancelling the first noise from the captured first voice signal, when determining that the first device and the second device are adjacent to each other and the capture of the first voice signal and the capture of the second voice signal overlap in time.

15. The method of claim 14, wherein the second location information includes a GPS coordinate of the second device, Wi-Fi access point information of the second device, cell information of a cellular network accessed by the second device, a location identification of the second device, and information of other devices detected by the second device.

16. The method of claim 14, wherein the second time information includes a start time of the second voice signal and an end time of the second voice signal.

17. The method of claim 14, further comprising:

calculating a correlation coefficient between the captured first voice signal and the captured second voice signal, and

determining whether the first voice signal and the second voice signal interfere with each other based at least in part on the calculated correlation coefficient.

18. The method of claim 14, further comprising:

cancelling from the captured second voice signal second noise which is associated with the first voice signal and estimating the second voice signal; and

translating the second voice signal into a second command for the second device.

19. The method of claim 18, further comprising:

transmitting the second command to the second device.

20. A computer-readable storage medium having stored thereon computer-executable instructions that, in response to execution, cause a speech recognition system to perform operations, comprising:

receiving, from a first device of a first user, first data including a first voice signal captured by the first device, first location information of the first device and first time information corresponding to the captured first voice signal;

receiving, from a second device of a second user, second data including a second voice signal captured by the second device, second location information of the second device and second time information corresponding to the captured second voice signal;

determining whether the first device and the second device are adjacent to each other based at least in part on the first location information and the second location information;

determining whether the capture of the first voice signal and the capture of the second voice signal overlap in time based at least in part on the first time information and the second time information, when determining that the first device and the second device are adjacent to each other;

cancelling first noise from the captured first voice signal when determining that the first device and the second device are adjacent to each other and the capture of the first voice signal and the capture of the second voice signal overlap in time, the first noise being associated with the second voice signal of the second user of the second device; and

translating the first voice signal into a first command for the first device.

21. The computer-readable storage medium of claim 20, wherein the first location information includes at least one of a GPS (Global Positioning System) coordinate of the first device, Wi-Fi access point information of the first device, cell information of a cellular network accessed by the first device, a location identification of the first device, and information of other devices detected by the first device, and

wherein the second location information includes a GPS coordinate of the second device, Wi-Fi access point information of the second device, cell information of a cellular network accessed by the second device, a location identification of the second device, and information of other devices detected by the second device.

22. The computer-readable storage medium of claim 20, wherein the first time information includes a start time of the first voice signal and an end time of the first voice signal, and

wherein the second time information includes a start time of the second voice signal and an end time of the second voice signal.

23. The computer-readable storage medium of claim 20, wherein the operations further comprise:

calculating a correlation coefficient between the captured first voice signal and the captured second voice signal, and

determining whether the first voice signal and the second voice signal interfere with each other based at least in part on the calculated correlation coefficient.

24. The computer-readable storage medium of claim 20, wherein the operations further comprise:

transmitting the first command to the first device.

25. The computer-readable storage medium of claim 20, wherein the operations further comprise:

cancelling from the captured second voice signal second noise associated with the first voice signal and estimating the second voice signal, when determining that the first device and the second device are adjacent to each other and the capture of the first voice signal and the capture of the second voice signal overlap in time; and

translating the second voice signal into a second command for the second device.

26. The computer-readable storage medium of claim 25, wherein the operations further comprise:

transmitting the second command to the second device.