System and method of spoken language understanding in human computer dialogs

• US 9,548,051 B2
• Filed: 02/03/2016
• Issued: 01/17/2017
• Est. Priority Date: 12/07/2001
• Status: Expired due to Fees

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property II LP (AT&T, Inc.)
• Inventors: Bangalore, Srinivas, Gupta, Narendra K., Rahim, Mazin G.
• Primary Examiner(s): PULLIAS, JESSE SCOTT
• Application Number:
  US15/014,070
• Publication Number:
  US 20160155439A1
• Time in Patent Office: 349 Days
• Field of Search: 704/231, 704/257, 704/251, 704/275, 704/9, 704/270.1, 704/1
• US Class Current: 1/1
• 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 speech recognizer output into a plurality of independent clauses; and

for an independent clause in the plurality of independent clauses;

identifying an object; and

recursively generating, via a processor, for each sub-independent clause within the independent clause, a semantic representation using the object.

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, wherein the object comprises a domain-dependent object.

4. The method of claim 1, further comprising:

identifying, in a domain-dependent manner, an action within the independent clause, wherein recursively generating the semantic representation further comprises using the action.

5. The method of claim 1, wherein while recursively generating the semantic representation, additional objects are extracted from the independent clause.

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

7. The method of claim 1, wherein recursively generating the semantic representation further comprises identifying relationships between a dialog act in the independent clause and the object.

8. The method of claim 1, wherein recursively generating the semantic representation further comprises filling in a data structure associated with a dialog act in the independent clause.

9. A system comprising:

a processor; and

a computer-readable storage device having instructions stored which, when executed by the processor, cause the processor to perform operations comprising;

partitioning speech recognizer output into a plurality of independent clauses; and

for an independent clause in the plurality of independent clauses;

identifying an object; and

recursively generating, via a processor, for each sub-independent clause within the independent clause, a semantic representation using the object.

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

11. The system of claim 9, wherein the object comprises a domain-dependent object.

12. The system of claim 9, further comprising:

identifying, in an domain-dependent manner, an action within the independent clause, wherein recursively generating the semantic representation further comprises using the action.

13. The system of claim 9, wherein while recursively generating the semantic representation, additional objects are extracted from the independent clause.

14. The system of claim 9, wherein identifying the object comprises using a domain specific classifier.

15. The system of claim 9, wherein recursively generating the semantic representation further comprises identifying relationships between a dialog act in the independent clause and the object.

16. The system of claim 9, wherein recursively generating the semantic representation further comprises filling in a data structure associated with a dialog act in the independent clause.

17. A non-transitory computer-readable storage device having instructions stored which, when executed by a processor, cause the processor to perform operations comprising:

partitioning speech recognizer output into a plurality of independent clauses; and

for an independent clause in the plurality of independent clauses;

identifying an object; and

recursively generating, via a processor, for each sub-independent clause within the independent clause, a semantic representation using the object.

18. The non-transitory computer-readable storage device of claim 17, wherein the semantic representation is used by a dialog manager in a spoken dialog system to determine a response to a user input.

19. The non-transitory computer-readable storage device of claim 17, wherein the object comprises a domain-dependent object.

20. The non-transitory computer-readable storage device of claim 17, further comprising:

identifying, in an domain-dependent manner, an action within the independent clause, wherein recursively generating the semantic representation further comprises using the action.