System, method and computer program product for a distributed speech recognition tuning platform

• US 20020138276A1
• Filed: 01/24/2001
• Published: 09/26/2002
• Est. Priority Date: 01/24/2001
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Bevocal Incorporated (Microsoft Corporation)
• Inventors: Damiba, Bertrand A.
• Granted Patent: US 7,383,187 B2
• Time in Patent Office: Days
• US Class Current: 704/270.1
• CPC Class Codes: G10L 15/06 Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice G10L15/14 takes precedence
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

A system, method and computer program product are provided for tuning a speech recognition process. Initially, a database of utterances is maintained. Thereafter, information associated with the utterances is collected utilizing a speech recognition process. Further, the utterances in the database are transmitted to a plurality of users utilizing a network. As such, transcriptions of the utterances in the database may be received from the users utilizing the network. In use, the speech recognition process may be tuned utilizing the information and the transcriptions.

Claims

1. A method for tuning a speech recognition process, comprising the steps of:

(a) maintaining a database of utterances;

(b) collecting information associated with the utterances in the database utilizing a speech recognition process;

(c) transmitting the utterances in the database to a plurality of users utilizing a network;

(d) receiving transcriptions of the utterances in the database from the users utilizing the network;

(e) tuning the speech recognition process utilizing the information and the transcriptions.

2. The method as recited in claim 1, wherein the network includes the Internet.

3. The method as recited in claim 2, wherein the transcriptions of the utterances are received from the users using a network browser.

4. The method as recited in claim 1, wherein the speech recognition process is tuned by performing experiments based on the information.

5. The method as recited in claim 4, wherein the information includes a recognition result.

6. A computer program product for tuning a speech recognition process, comprising:

(a) computer code for maintaining a database of utterances;

(b) computer code for collecting information associated with the utterances in the database utilizing a speech recognition process;

(c) computer code for transmitting the utterances in the database to a plurality of users utilizing a network;

(d) computer code for receiving transcriptions of the utterances in the database from the users utilizing the network;

(e) computer code for tuning the speech recognition process utilizing the information and the transcriptions.

7. The computer program product as recited in claim 6, wherein the network includes the Internet.

8. The computer program product as recited in claim 7, wherein the transcriptions of the utterances are received from the users using a network browser.

9. The computer program product as recited in claim 6, wherein the speech recognition process is tuned by performing experiments based on the information.

10. The computer program product as recited in claim 9, wherein the information includes a recognition result.

11. A system for tuning a speech recognition process, comprising:

(a) logic for maintaining a database of utterances;

(b) logic for collecting information associated with the utterances in the database utilizing a speech recognition process;

(c) logic for transmitting the utterances in the database to a plurality of users utilizing a network;

(d) logic for receiving transcriptions of the utterances in the database from the users utilizing the network;

(e) logic for tuning the speech recognition process utilizing the information and the transcriptions.

12. The system as recited in claim 11, wherein the network includes the Internet.

13. The system as recited in claim 12, wherein the transcriptions of the utterances are received from the users using a network browser.

14. The system as recited in claim 11, wherein the speech recognition process is tuned by performing experiments based on the information.

15. The system as recited in claim 14, wherein the information includes a recognition result.