SYSTEM AND METHOD OF PERFORMING AUTOMATIC SPEECH RECOGNITION USING LOCAL PRIVATE DATA

• US 20150120288A1
• Filed: 10/29/2013
• Published: 04/30/2015
• Est. Priority Date: 10/29/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: JOHNSTON, Michael J., THOMSON, David, RANGARAJAN SRIDHAR, Vivek Kumar
• Granted Patent: US 9,666,188 B2
• Time in Patent Office: Days
• US Class Current: 704/231
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A method of providing hybrid speech recognition between a local embedded speech recognition system and a remote speech recognition system relates to receiving speech from a user at a device communicating with a remote speech recognition system. The system recognizes a first part of speech by performing a first recognition of the first part of the speech with the embedded speech recognition system that accesses private user data, wherein the private user data is not available to the remote speech recognition system. The system recognizes the second part of the speech by performing a second recognition of the second part of the speech with the remote speech recognition system. The final recognition result is a combination of these two recognition processes. The private data can be such local information as a user location, a playlist, frequently dialed numbers or texted people, user contact list information, and so forth.

Claims

1. A method comprising:

receiving speech from a user at a device communicating with a remote speech recognition system, wherein the device comprises an embedded speech recognition system that accesses private user data; and

recognizing a part of the speech with the embedded speech recognition system by accessing the private user data, wherein the private user data is not available to the remote speech recognition system.

2. The method of claim 1, wherein the part of the speech comprises data associated with an identification of a person rather than words spoken by the user.

3. The method of claim 1, wherein recognizing the part of the speech further comprises:

receiving from the remote speech recognition system an entity associated with options to use for speech recognition;

evaluating the entity at the embedded speech recognition system in view of the private user data to yield an evaluation; and

selecting a recognition result based on the evaluation.

4. The method of claim 3, further comprising:

receiving data from the remote speech recognition system which comprises a placeholder in place of an entity.

5. The method of claim 4, wherein the placeholder is one of a garbage model, phonemic language model, a language model based on a standard list, and a language model built on the private data.

6. The method of claim 5, wherein individual names from a user contact list are from the private user data are not extractable from the language model.

7. The method of claim 1, wherein the private user data comprises one of data in a user contact list, frequently dialed phone numbers, frequently used texted names, data associated with a user location, data associated with a playlist, user history, and multiple hypothesis associated with private information.

8. A computer-readable storage device storing instructions which, when executed by the processor, cause the processor to perform operations comprising:

receiving speech from a user at a device communicating with a remote speech recognition system, wherein the device comprises an embedded speech recognition system that accesses private user data; and

recognizing a part of the speech with the embedded speech recognition system by accessing the private user data, wherein the private user data is not available to the remote speech recognition system.

9. The computer-readable storage device of claim 8, wherein the part of the speech comprises data associated with an identification of a person rather than words spoken by the user.

10. The computer-readable storage device of claim 8, wherein recognizing the part of the speech further comprises:

receiving from the remote speech recognition system an entity associated with options to use for speech recognition;

evaluating the entity at the embedded speech recognition system in view of the private user data to yield an evaluation; and

selecting a recognition result based on the evaluation.

11. The computer-readable storage device of claim 10, further storing instructions which, when executed by the processor, cause the processor to perform an operation comprising:

receiving data from the remote speech recognition system which comprises a placeholder in place of a entity.

12. The computer-readable storage device of claim 11, wherein the placeholder is one of a garbage model, phonemic language model, a language model based on a standard list, and a language model built on the private data.

13. The computer-readable storage device of claim 12, wherein individual names from a user contact list are from the private user data are not extractable from the language model.

14. The computer-readable storage device of claim 8, wherein the private user data comprises one of data in a user contact list, frequently dialed phone numbers, frequently used texted names, data associated with a user location, data associated with a playlist, user history, and multiple hypothesis associated with private information.

15. A system comprising:

a processor; and

computer-readable storage medium storing instructions which, when executed by the processor, cause the processor to perform operations comprising;

receiving speech from a user at a device communicating with a remote speech recognition system, wherein the device comprises an embedded speech recognition system that accesses private user data; and

recognizing a part of the speech with the embedded speech recognition system by accessing the private user data, wherein the private user data is not available to the remote speech recognition system.

16. The system of claim 15, wherein the part of the speech comprises data associated with an identification of a person rather than words spoken by the user.

17. The system of claim 16, wherein recognizing the part of the speech further comprises:

receiving from the remote speech recognition system an entity associated with options to use for speech recognition;

evaluating the entity at the embedded speech recognition system in view of the private user data to yield an evaluation; and

selecting a recognition result based on the evaluation.

18. The system of claim 17, wherein the computer-readable storage medium stores further instructions which, when executed by the processor, cause the processor to perform an operation comprising:

receiving data from the remote speech recognition system which includes a placeholder in place of a entity.

19. The system of claim 18, wherein the placeholder is one of a garbage model, phonemic language model, a language model based on a standard list, and a language model built on the private data.

20. The system of claim 15, wherein the private user data comprises one of data in a user contact list, frequently dialed phone numbers, frequently used texted names, data associated with a user location, user history, data associated with a playlist, and multiple hypothesis associated with private information.