AUTOMATIC GRAMMAR TUNING USING STATISTICAL LANGUAGE MODEL GENERATION

• US 20080052076A1
• Filed: 08/22/2006
• Published: 02/28/2008
• Est. Priority Date: 08/22/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: METZ, BRENT D.
• Granted Patent: US 8,346,555 B2
• Time in Patent Office: Days
• US Class Current: 704/257
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/197 Probabilistic grammars, e.g. word n-grams

Abstract

The present invention discloses a speech processing solution that utilizes an original speech recognition grammar in a speech recognition system to perform speech recognition operations for multiple recognition instances. Instance data associated with the recognition operations can be stored. A replacement grammar can be automatically generated from the stored instance data, where the replacement grammar is a statistical language model grammar. The original speech recognition grammar, which can be a grammar-based language model grammar or a statistical language model grammar, can be selectively replaced with the replacement grammar. For example when tested performance for the replacement grammar is better than that for the original grammar, the replacement grammar can replace the original grammar.

Claims

1. A speech processing method comprising:

utilizing an original speech recognition grammar in a speech recognition system to perform speech recognition operations for a plurality of recognition instances;

storing instance data associated with the recognition operations;

automatically generating a replacement grammar from the stored instance data, wherein the replacement grammar is a statistical language model grammar; and

selectively replacing the original speech recognition grammar in the speech recognition system with the replacement grammar.

2. The method of claim 1 wherein the original speech recognition grammar is a grammar-based language model grammar.

3. The method of claim 1, wherein the original speech recognition grammar is written in a grammar format specification language selected from a group of languages consisting of a NUANCE Grammar Specification Language (GSL), a Speech Recognition Grammar Specification (SRGS) compliant language and a JAVA Speech Grammar Format (JSGF) compliant language.

4. The method of claim 1 wherein the original speech recognition grammar is a statistical language model grammar.

5. The method of claim 1, further comprising:

comparing a performance of the original speech recognition grammar against the replacement grammar, wherein the replacement step is contingent upon results of the comparing step, and wherein the replacing step selectively occurs when the performance of the replacement grammar favorably compares to the performance of the original speech recognition grammar.

6. The method of claim 5, further comprising:

automatically and dynamically performing the comparing step and the replacing step when the performance of the replacement grammar favorably compares to the performance of the original speech recognition grammar.

7. The method of claim 5, further comprising:

presenting an administrator of the speech recognition system with an option to replace the original speech recognition grammar with the replacement grammar wherein the replacement step is contingent upon response provided for the option.

8. The method of claim 1, wherein the speech recognition grammar comprises a plurality of grammars, which include said original speech recognition grammar.

9. The method of claim 8, wherein the original speech recognition grammar and the replacement grammar are context dependent grammars.

10. The method of claim 8, wherein the original speech recognition grammar and the replacement grammar are speaker dependent grammars.

11. The method of claim 1, wherein the original speech recognition grammar and the replacement grammar are context-independent grammars.

12. The method of claim 1, wherein said steps of claim 1 are steps performed by at least one machine in accordance with at least one computer program having a plurality of code sections that are executable by the at least one machine.

13. A method for tuning speech recognition grammars comprising:

performing a plurality of speech-to-text operations using an original speech recognition grammar, wherein the original speech recognition grammar is a grammar-based language model grammar.

recording, recognition instances associated with the speech-to-text operations;

automatically creating a set of words and phrases from the recorded recognition data;

automatically generating a replacement grammar from the set of words and phrases, wherein the replacement grammar is a statistical language model grammar; and

replacing the original speech recognition grammar with the replacement grammar.

14. The method of claim 13, further comprising:

automatically weighing the words and phrases based upon the recorded recognition data.

15. The method of claim 13, further comprising:

executing replacement speech-to-text operations based upon performance testing input using the replacement grammar.

generating replacement performance metrics based upon the replacement speech-to-text operations;

ascertaining original performance metrics for the speech-to-text operations based upon the performance testing input using the original speech recognition grammar; and

comparing the replacement performance metrics and the original performance metrics, wherein the replacing step is contingent upon results of the comparing step, wherein the replacing step selectively occurs when the replacement performance metrics favorably compare to the original performance metrics.

16. A speech recognition system comprising:

a language model processor configured to utilize an original speech recognition grammar in performing speech recognition operations;

a log data store configured to store speech instance data associated with the speech recognition operations;

a statistical language model generator configured to automatically generate a replacement grammar front the speech instance data; and

a grammar swapper configured to selectively replace the original speech recognition grammar with the speech replacement grammar.

17. The system of claim 16, wherein the original speech recognition grammar is a grammar-based language model grammar.

18. The system of claim 16, wherein the original speech recognition grammar is written in a grammar format specification language selected from a group of languages consisting of a NUANCE Grammar Specification Language (GSL), a Speech Recognition Grammar Specification (SRGS) compliant language and a JAVA Speech Grammar Format (JSGF) compliant language.

19. The system of claim 16, wherein the original speech recognition grammar and the replacement grammar are context dependent grammars.

20. The system of claim 16, further comprising:

a performance analyzer configured to compare a performance of original speech recognition grammar with a performance of the replacement grammar, wherein actions taken by the grammar swapper are contingent upon results of comparisons performed by the performance analyzer.