SYSTEM AND METHOD FOR INCREASING ACCURACY OF SEARCHES BASED ON COMMUNITIES OF INTEREST

• US 20090112600A1
• Filed: 10/31/2007
• Published: 04/30/2009
• Est. Priority Date: 10/31/2007
• Status: Active Grant

• Current Assignee: AT&T Intellectual Property II LP (AT&T, Inc.)
• Original Assignee: AT&T Corporation (AT&T, Inc.)
• Inventors: Gilbert, Mazin, Volinsky, Christopher
• Granted Patent: US 7,966,171 B2
• Time in Patent Office: Days
• US Class Current: 704/270.100
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/193 Formal grammars, e.g. finite state automata, context free grammars or word networks
• CPC Class Codes: G10L 2015/228 of application context

Abstract

Disclosed are systems, methods and computer-readable media for using a local communication network to generate a speech model. The method includes retrieving for an individual a list of numbers in a calling history, identifying a local neighborhood associated with each number in the calling history, truncating the local neighborhood associated with each number based on the at least one parameter, retrieving a local communication network associated with each number in the calling history and each phone number in the local neighborhood, and creating a language model for the individual based on the retrieved local communication network. The generated language model may be used for improved automatic speech recognition for audible searches as well as other modules in a spoken dialog system.

Claims

1. A method of using a local communication network to generate a speech recognition model, the method comprising:

retrieving for an individual a list of numbers in a calling history;

identifying a local neighborhood associated with each number in the calling history;

truncating the local neighborhood associated with each number based on at least one parameter;

retrieving a local communication network associated with each number in the calling history and each phone number associated with a local neighborhood; and

creating a language model for the individual based on the retrieved local communication network.

2. The method of claim 1, wherein the local neighborhood further comprises at least one of a corresponding entity, a business name, a person'"'"'s name and data associated with the corresponding entity, business name or person.

3. The method of claim 1, further comprising implementing the created language model in a spoken dialog system.

4. The method of claim 1, further comprising using the local neighborhood to create a punctuation dictionary and a language model.

5. The method of claim 1, wherein the language model is one of a deterministic model or a stochastic model.

6. The method of claim 1, further comprising dynamically modifying the created language model based on additional data.

7. The method of claim 1, further comprising implementing the created language model in a spoken dialog system at a time based on a history of the user and to anticipate user input.

8. A speech recognition model created by a method comprising:

retrieving for an individual a list of numbers in a calling history;

identifying a local neighborhood associated with each number in the calling history;

truncating the local neighborhood associated with each number based on at least one parameter;

retrieving a local communication network associated with each number in the calling history and each phone number associated with a local neighborhood; and

creating a language model for the individual based on the retrieved local communication network.

9. The speech recognition model of claim 8, wherein the local neighborhood further comprises at least one of a corresponding entity, a business name, a person'"'"'s name and data associated with the corresponding entity, business name or person.

10. The speech recognition model of claim 8, wherein the method used to create the speech recognition further comprises using the local neighborhood to create a punctuation dictionary and a language model.

11. The speech recognition model of claim 8, wherein the language model is one of a deterministic model or a stochastic model.

12. The speech recognition model of claim 8, further comprising dynamically modifying the created language model based on additional data.

13. The speech recognition model of claim 8, further comprising implementing the created language model in a spoken dialog system at a time based on a history of the user and to anticipate user input.

15. The spoken dialog system of claim 12, wherein the local neighborhood further comprises at least one of a corresponding entity, a business name, a person'"'"'s name and data associated with the corresponding entity, business name or person.

16. The spoken dialog system of claim 12, further comprising using the local neighborhood to create a punctuation dictionary and a language model.

17. The spoken dialog system of claim 12, wherein the language model is one of a deterministic model or a stochastic model.

18. The spoken dialog system of claim 12, further comprising dynamically modifying the created language model based on additional data.

19. The spoken dialog system of claim 12, further comprising implementing the created language model in a spoken dialog system at a time based on a history of the user and to anticipate user input.

14. A spoken dialog system having a speech model for use by at least the automatic speech recognition module, the speech model generated by a method comprising:

retrieving for an individual a list of numbers in a calling history;

identifying a local neighborhood associated with each number in the calling history;

truncating the local neighborhood associated with each number based on at least one parameter;

retrieving a local communication network associated with each number in the calling history and each phone number associated with a local neighborhood; and

creating a language model for the individual based on the retrieved local communication network.