Automatic speech recognition with dynamic grammar rules

• US 8,670,987 B2
• Filed: 03/20/2007
• Issued: 03/11/2014
• Est. Priority Date: 03/20/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Bergl, Vladimir, Cross, Charles W., Jania, Frank
• Primary Examiner(s): He, Jialong
• Application Number:
  US11/688,334
• Publication Number:
  US 20080235022A1
• Time in Patent Office: 2,548 Days
• Field of Search: 704/270, 704/275
• US Class Current: 704/275
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/32 Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems
• CPC Class Codes: G10L 2015/228 of application context

Abstract

Automatic speech recognition implemented with a speech recognition grammar of a multimodal application in an ASR engine, the multimodal application operating on a multimodal device supporting multiple modes of user interaction including a voice mode, the multimodal application operatively coupled to the ASR engine, including: matching by the ASR engine at least one static rule of the speech recognition grammar with at least one word of a voice utterance, yielding a matched value, the matched value specified by the grammar to be required for processing of a dynamic rule of the grammar; and dynamically defining at run time the dynamic rule of the grammar as a new static rule in dependence upon the matched value, the dynamic rule comprising a rule that is specified by the grammar as a rule that is not to be processed by the ASR until after the at least one static rule has been matched.

Claims

1. A method for use in connection with automatic speech recognition, the method implemented at least in part by using an automatic speech recognition engine, the method comprising:

matching, by using at least one processor, at least one static rule of a speech recognition grammar with at least one word of a voice utterance yielding at least one matched value, the at least one static rule referenced in a parameter list of a dynamic rule of the speech recognition grammar, the parameter list occurring after the dynamic rule in the speech recognition grammar; and

in response to the matching, obtaining a new static rule by processing the dynamic rule at least in part by using the at least one matched value,

wherein the dynamic rule is not processed by the automatic speech recognition engine until after the at least one static rule has been matched.

2. The method of claim 1, further comprising treating the dynamic rule as a rule that matches any speech in the voice utterance until a next rule match, a next token match, or the end of processing of the voice utterance.

3. The method of claim 1, wherein the speech recognition grammar includes a definition of the dynamic rule that identifies a computer resource configured to dynamically define at run time the dynamic rule of the grammar as a new static rule in dependence upon the matched value.

4. The method of claim 1, wherein processing the dynamic rule comprises expanding a definition of the dynamic rule with the at least one matched value of the at least one static rule.

5. The method of claim 1, wherein the speech recognition grammar comprises a plurality of dynamic rule references to the dynamic rule, each of the plurality of dynamic rule references including a static rule parameter list comprising one or more static rule references specifying one or more static rules required to be matched before processing the dynamic rule.

6. An apparatus, comprising:

at least one computer processor configured to perform, at least in part by using an automatic speech recognition engine, acts of;

matching at least one static rule of a speech recognition grammar with at least one word of a voice utterance yielding at least one matched value, the at least one static rule referenced in a parameter list of a dynamic rule of the speech recognition grammar, the parameter list occurring after the dynamic rule in the speech recognition grammar; and

in response to the matching, obtaining a new static rule by processing the dynamic rule at least in part by using the at least one matched value,

wherein the dynamic rule is not processed by the automatic speech recognition engine until after the at least one static rule has been matched.

7. The apparatus of claim 6, wherein the at least one computer processor is further configured to treat the dynamic rule as a rule that matches any speech in the voice utterance until a next rule match, a next token match, or the end of processing of the voice utterance.

8. The apparatus of claim 6, wherein the speech recognition grammar includes a definition of the dynamic rule that identifies a computer resource configured to dynamically define at run time the dynamic rule of the speech recognition grammar as a new static rule in dependence upon the matched value.

9. The apparatus of claim 6, wherein processing the dynamic rule comprises expanding a definition of the dynamic rule with the at least one matched value of the at least one static rule.

10. The apparatus of claim 6, wherein the speech recognition grammar comprises a plurality of dynamic rule references to the dynamic rule, each of the plurality of dynamic rule references including a static rule parameter list comprising one or more static rule references specifying one or more static rules required to be matched before processing the dynamic rule.

11. At least one computer readable recordable storage medium storing computer program instructions that, when executed by at least one processor, cause the at least one processor to perform a method for use in connection with automatic speech recognition, the method implemented at least in part by using an automatic speech recognition engine, the method comprising:

matching at least one static rule of a speech recognition grammar with at least one word of a voice utterance yielding at least one matched value, the at least one static rule referenced in a parameter list of a dynamic rule of the speech recognition grammar, the parameter list occurring after the dynamic rule in the speech recognition grammar; and

in response to the matching, obtaining a new static rule by processing the dynamic rule at least in part by using the at least one matched value,

wherein the dynamic rule is not processed by the automatic speech recognition engine until after the at least one static rule has been matched.

12. The at least one computer readable recordable storage medium of claim 11, wherein the method further comprises treating the dynamic rule as a rule that matches any speech in the voice utterance until a next rule match, a next token match, or the end of processing of the voice utterance.

13. The at least one computer readable recordable storage medium of claim 11, wherein the speech recognition grammar includes a definition of the dynamic rule that identifies a computer resource configured to dynamically define at run time the dynamic rule of the speech recognition grammar as a new static rule in dependence upon the matched value.

14. The at least one computer readable recordable storage medium of claim 11, wherein processing the dynamic rule comprises expanding a definition of the dynamic rule with the at least one matched value of the at least one static rule.

15. The at least one computer readable recordable storage medium of claim 11, wherein the speech recognition grammar comprises a plurality of dynamic rule references to the dynamic rule, each of the plurality of dynamic rule references including a static rule parameter list comprising one or more static rule references specifying one or more static rules required to be matched before processing the dynamic rule.