IN-VEHICLE CIRCUMSTANTIAL SPEECH RECOGNITION

• US 20090164216A1
• Filed: 12/21/2007
• Published: 06/25/2009
• Est. Priority Date: 12/21/2007
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: General Motors Corporation (Motors Liquidation Co. GUC Trust)
• Inventors: Abeska, Edward J., Chengalvarayan, Rathinavelu, Grost, Timothy J.
• Granted Patent: US 8,423,362 B2
• Time in Patent Office: Days
• US Class Current: 704/251
• CPC Class Codes: B60R 16/0373 Voice control in general G10L

Abstract

A method of circumstantial speech recognition in a vehicle. A plurality of parameters associated with a plurality of vehicle functions are monitored as an indication of current vehicle circumstances. At least one vehicle function is identified as a candidate for user-intended ASR control based on user interaction with the vehicle. The identified vehicle function is then used to disambiguate between potential commands contained in speech received from the user.

Claims

1. A method of circumstantial speech recognition in a vehicle, the method comprising the steps of:

monitoring a plurality of parameters associated with a plurality of vehicle functions as an indication of current vehicle circumstances; and

identifying at least one vehicle function as a candidate for user-intended ASR control when at least one of the monitored plurality of parameters associated with at least one of the plurality of vehicle functions meets predetermined criteria.

2. The method set forth in claim 1, further comprising the steps of:

receiving user speech;

pre-processing the user speech;

decoding the pre-processed speech into recognized speech data; and

controlling the identified at least one vehicle function using the recognized speech data.

3. The method set forth in claim 2, wherein the pre-processed speech is decoded by applying at least one speech recognition model associated with at least one of the plurality of vehicle functions.

4. The method set forth in claim 1, wherein the plurality of parameters includes a user initiated event.

5. The method set forth in claim 4, wherein the user initiated event includes connection of an external device to a vehicle device.

6. A method of circumstantial speech recognition in a vehicle based on user interactivity with the vehicle, the method comprising the steps of:

receiving speech from a user;

pre-processing the received speech;

decoding the pre-processed speech into recognized speech data;

monitoring a plurality of vehicle devices for interaction by the user;

identifying a vehicle device for user-intended ASR control if the user has interacted with the one vehicle device within a predetermined elapsed time; and

controlling the identified vehicle device using the recognized speech data.

7. The method set forth in claim 6, wherein the pre-processed speech is decoded by applying at least one speech recognition model associated with the identified vehicle device.

8. The method set forth in claim 6, wherein the user interaction includes connection of an external device to a vehicle device.

9. A method of circumstantial speech recognition in a vehicle based on user interactivity with the vehicle, the method comprising the steps of:

monitoring a plurality of vehicle devices for interaction by a user;

identifying a vehicle device for user-intended ASR control based on user interaction with the vehicle device;

receiving speech from the user; and

disambiguating between two or more possible commands contained in the speech based at least in part on the identified vehicle device.

10. The method set forth in claim 9, wherein the disambiguating step further comprises recognizing the speech using a speech recognition model selected based at least in part on the identified vehicle device.

11. The method set forth in claim 9, wherein the disambiguating step further comprises recognizing the speech using an ASR system and thereafter determining which of two or more potential commands is intended by the user based at least in part on the identified vehicle device.

12. The method set forth in claim 9, further comprising the step of controlling the identified vehicle device using the recognized speech data.

13. The method set forth in claim 9, further comprising the step of pre-processing the received speech before the disambiguating step.

14. The method set forth in claim 9, wherein the monitoring step further comprises monitoring for connection of an external device to one of the vehicle devices.