System and method of spoken language understanding in human computer dialogs

• US 8,386,262 B2
• Filed: 05/25/2012
• Issued: 02/26/2013
• Est. Priority Date: 12/07/2001
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property II LP (AT&T, Inc.)
• Inventors: Rahim, Mazin G., Bangalore, Srinivas, Gupta, Narendra K.
• Primary Examiner(s): PULLIAS, JESSE SCOTT
• Application Number:
  US13/481,031
• Publication Number:
  US 20120239383A1
• Time in Patent Office: 277 Days
• Field of Search: 704/1, 704/9, 704/231, 704/251, 704/257, 704/270.1, 704/275
• US Class Current: 704/270.1
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A system and method are disclosed that improve automatic speech recognition in a spoken dialog system. The method comprises partitioning speech recognizer output into self-contained clauses, identifying a dialog act in each of the self-contained clauses, qualifying dialog acts by identifying a current domain object and/or a current domain action, and determining whether further qualification is possible for the current domain object and/or current domain action. If further qualification is possible, then the method comprises identifying another domain action and/or another domain object associated with the current domain object and/or current domain action, reassigning the another domain action and/or another domain object as the current domain action and/or current domain object and then recursively qualifying the new current domain action and/or current object. This process continues until nothing is left to qualify.

Claims

1. A method comprising:

partitioning, via a processor, a speech recognizer output into independent clauses;

identifying, independent of domain, a dialog act for each of the independent clauses;

identifying, dependent on domain, an object within each of the independent clauses;

generating a semantic representation using the dialog act and the object; and

recursively extending the semantic representation by qualifying the object in each of the independent clauses.

2. The method of claim 1, wherein the semantic representation is used by a dialog manager in a spoken dialog system to determine a response to a user input.

3. The method of claim 1, further comprising:

identifying, dependent on domain, an action within each of the independent clauses, wherein generating the semantic representation further comprises using the action.

4. The method of claim 1, wherein qualifying the object in each of the independent clauses comprises extracting additional objects from each of the independent clauses.

5. The method of claim 1, wherein identifying the object comprises using a domain specific classifier.

6. The method of claim 1, wherein generating the semantic representation further comprises identifying relationships between the dialog act and the object.

7. The method of claim 1, wherein generating the semantic representation further comprises filling in a predefined data structure associated with the dialog act.

8. A system comprising:

a processor; and

a computer-readable storage medium having instructions stored which, when executed on the processor, perform a method comprising;

partitioning a speech recognizer output into independent clauses;

identifying, independent of domain, a dialog act for each of the independent clauses;

identifying, dependent on domain, an object within each of the independent clauses;

generating a semantic representation using the dialog act and the object; and

recursively extending the semantic representation by qualifying the object in each of the independent clauses.

9. The system of claim 8, wherein the semantic representation is used by a dialog manager in a spoken dialog system to determine a response to a user input.

10. The system of claim 8, the computer-readable storage medium having additional instructions stored which, when executed on the processor, perform a method comprising:

identifying, dependent on domain, an action within each of the independent clauses, wherein generating the semantic representation further comprises using the action.

11. The system of claim 8, wherein qualifying the object in each of the independent clauses comprises extracting additional objects from each of the independent clauses.

12. The system of claim 8, wherein identifying the object comprises using a domain specific classifier.

13. The system of claim 8, wherein generating the semantic representation further comprises identifying relationships between the dialog act and the object.

14. The system of claim 8, wherein generating the semantic representation further comprises filling in a predefined data structure associated with the dialog act.

15. A computer-readable storage device having instructions stored which, when executed on a computing device, perform a method comprising:

partitioning a speech recognizer output into independent clauses;

identifying, independent of domain, a dialog act for each of the independent clauses;

identifying, dependent on domain, an object within each of the independent clauses;

generating a semantic representation using the dialog act and the object; and

recursively extending the semantic representation by qualifying the object in each of the independent clauses.

16. The computer-readable storage device of claim 15, wherein the semantic representation is used by a dialog manager in a spoken dialog system to determine a response to a user input.

17. The computer-readable storage device of claim 15, the computer-readable storage device having additional instructions stored which, when executed on the computing device, perform a method comprising identifying, dependent on domain, an action within each of the independent clauses, wherein generating the semantic representation further comprises using the action.

18. The computer-readable storage device of claim 15, wherein qualifying the object in each of the independent clauses comprises extracting additional objects from each of the independent clauses.

19. The computer-readable storage device of claim 15, wherein identifying the object comprises using a domain specific classifier.

20. The computer-readable storage device of claim 15, wherein generating the semantic representation further comprises identifying relationships between the dialog act and the object.