Method and system for improved speech recognition by degrading utterance pronunciations

• US 20070038454A1
• Filed: 08/10/2005
• Published: 02/15/2007
• Est. Priority Date: 08/10/2005
• Status: Active Grant

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Ruback, Harvey, Eckhart, John
• Granted Patent: US 7,983,914 B2
• Time in Patent Office: Days
• US Class Current: 704/260
• CPC Class Codes: G10L 15/063 Training

Abstract

A speech recognition system (10) or method (20) can include a speech input device and a processor (14) coupled to the speech input. The processor can be programmed to identify (22) a plurality of words that are members of confusable pairs of words where each pair includes a target word and a substituted word. The processor can degrade (24) a pronunciation of the substituted word to provide a worse pronunciation of the substituted word. The processor can further compare (28) the pronunciation of the target word with the worse pronunciation to the substituted word. The processor can be further programmed to reduce (26) confusion between the substituted word and other words in a recognition grammar of the speech recognition engine and can also narrow the scope within which the substituted word is recognized.

Claims

1. In a speech recognition engine, a method of improving overall system recognition performance, comprising the steps of:

identifying a plurality of words that are members of confusable pairs of words, each pair comprising of a target word and a substituted word; and

degrading a pronunciation of the substituted word to provide worse pronunciation of the substituted word.

2. The method of claim 1, wherein the method further comprises the step of comparing the pronunciation of the target word with the worse pronunciation to the substituted word.

3. The method of claim 1, wherein the step of degrading the pronunciation comprises providing a slightly worse pronunciation to the substituted word.

4. The method of claim 1, wherein the method further comprises the step of causing the substituted word to be less favorable overall.

5. The method of claim 1, wherein the method reduces confusion between the substituted word and other words in a recognition grammar of the speech recognition engine.

6. The method of claim 1, wherein the method further comprises the step of narrowing the scope within which the substituted word is recognized.

7. The method of claim 6, wherein the method further comprises the step of reducing confusion between the substituted word and words in a recognition grammar other than the target word.

8. A speech recognition system, comprising:

a speech input

a processor coupled to the speech input, wherein the processor is programmed to;

identify a plurality of words that are members of confusable pairs of words, each pair comprising of a target word and a substituted word; and

degrade a pronunciation of the substituted word to provide worse pronunciation of the substituted word.

9. The speech recognition system of claim 8, wherein the processor is further programmed to compare the pronunciation of the target word with the worse pronunciation to the substituted word.

10. The speech recognition system of claim 8, wherein the processor degrades the pronunciation by providing a slightly worse pronunciation to the substituted word.

11. The speech recognition system of claim 8, wherein the processor is further programmed to cause the substituted word to be less favorable overall.

12. The speech recognition system of claim 8, wherein the processor is further programmed to reduce confusion between the substituted word and other words in a recognition grammar of the speech recognition engine.

13. The speech recognition system of claim 8, wherein the processor is further programmed to narrow the scope within which the substituted word is recognized.

14. The speech recognition system of claim 13, wherein the processor is further programmed to reduce confusion between the substituted word and words in a recognition grammar other than the target word.

15. A machine-readable storage, having stored thereon a computer program having a plurality of code sections executable by a machine for causing the machine to perform the steps of:

identifying a plurality of words that are members of confusable pairs of words, each pair comprising of a target word and a substituted word; and

degrading a pronunciation of the substituted word to provide worse pronunciation of the substituted word.

16. The machine readable storage of claim 15, wherein the computer program further comprises a plurality of code section for causing a machine to compare the pronunciation of the target word with the worse pronunciation to the substituted word.

17. The machine readable storage of claim 15, wherein the computer program degrades the pronunciation by providing a slightly worse pronunciation to the substituted word.

18. The machine readable storage of claim 15, wherein the computer program further comprises a plurality of code sections for causing the substituted word to be less favorable overall.

19. The machine readable storage of claim 15, wherein the computer program further comprises a plurality of code sections causing the machine to narrow the scope within which the substituted word is recognized.

20. The machine readable storage of claim 19, wherein the computer program further comprises a plurality of code sections for reducing confusion between the substituted word and words in a recognition grammar other than the target word.