Method and system for speech recognition using phonetically similar word alternatives

• US 20020173956A1
• Filed: 05/16/2001
• Published: 11/21/2002
• Est. Priority Date: 05/16/2001
• Status: Active Grant

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Hartley, Matthew W., Reich, David E.
• Granted Patent: US 6,910,012 B2
• Time in Patent Office: Days
• US Class Current: 704/231
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 2015/0636 Threshold criteria for the updating

Abstract

A method for performing speech recognition can include the steps of providing a grammar including entries comprising a parent word and a pseudo word being substantially phonetically equivalent to the parent word. The grammar can provide a translation from the pseudo word to the parent word. The parent word can be received as speech and the speech can be compared to the grammar entries. Additionally, the speech can be matched to the pseudo word and the pseudo word can be translated to the parent word.

Claims

1. A method for performing speech recognition comprising:

providing a grammar including entries comprising a parent word and a pseudo word being phonetically similar to said parent word, said grammar providing a translation from said pseudo word to said parent word;

receiving said parent word as speech;

comparing said speech to said grammar entries; and

selecting said pseudo word as a match for said speech and translating said pseudo word to said parent word.

2. The method of claim 1, said comparing step comprising:

generating a confidence score for said parent word and said pseudo word.

3. The method of claim 1, said matching step further comprising:

identifying said pseudo word having a highest confidence score.

4. The method of claim 1, wherein said parent word is a character.

5. A method for performing speech recognition comprising:

providing a spelling grammar including one or more entries comprising a parent letter and a pseudo word being phonetically similar to said parent letter, said spelling grammar providing a translation from said pseudo word to said parent letter;

receiving said parent letter as speech;

comparing said speech to said grammar entries; and

selecting said pseudo word as a match for said speech and translating said pseudo word to said parent letter.

6. The method of claim 5, said comparing step comprising:

generating a confidence score for said parent letter and said pseudo word.

7. The method of claim 5, said matching step further comprising:

identifying said pseudo word having a highest confidence score.

8. The method of claim 5, wherein said spelling grammar contains a pseudo word corresponding to each letter of the alphabet.

9. A method for performing speech recognition comprising:

providing a grammar including entries comprising a parent word and a pseudo word being substantially phonetically equivalent to said parent word, said grammar providing a translation from said pseudo word to said parent word;

receiving said parent word as speech;

comparing said speech to said grammar entries; and

selecting said pseudo word as a match for said speech and translating said pseudo word to said parent word.

10. A method for performing speech recognition comprising:

providing a spelling grammar including one or more entries comprising a parent letter and a pseudo word being substantially phonetically equivalent to said parent letter, said spelling grammar providing a translation from said pseudo word to said parent letter;

receiving said parent letter as speech;

comparing said speech to said grammar entries; and

selecting said pseudo word as a match for said speech and translating said pseudo word to said parent letter.

11. A speech recognition system comprising a speech recognition engine, said speech recognition engine including a grammar including a parent word and a pseudo word being substantially phonetically equivalent to said parent word, said grammar providing a translation from said pseudo word to said parent word.

12. The system of claim 11, wherein said parent word is a character.

13. The system of claim 11, wherein said grammar is a spelling grammar.

15. The machine readable storage of claim 14, said comparing step comprising:

generating a confidence score for said parent word and said pseudo word.

16. The machine readable storage of claim 14, said matching step further comprising:

identifying said pseudo word having a highest confidence score.

17. The machine readable storage of claim 14, wherein said parent word is a character.

19. The machine readable storage of claim 18, said comparing step comprising:

generating a confidence score for said parent letter and said pseudo word.

20. The machine readable storage of claim 18, said matching step further comprising:

identifying said pseudo word having a highest confidence score.

21. The machine readable storage of claim 18, wherein said spelling grammar contains a pseudo word corresponding to each letter of the alphabet.

14. 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:

providing a grammar including entries comprising a parent word and a pseudo word being phonetically similar to said parent word, said grammar providing a translation from said pseudo word to said parent word;

receiving said parent word as speech;

comparing said speech to said grammar entries; and

selecting said pseudo word as a match for said speech and translating said pseudo word to said parent word.

18. 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:

providing a spelling grammar including one or more entries comprising a parent letter and a pseudo word being phonetically similar to said parent letter, said spelling grammar providing a translation from said pseudo word to said parent letter;

receiving said parent letter as speech;

comparing said speech to said grammar entries; and

selecting said pseudo word as a match for said speech and translating said pseudo word to said parent letter.

22. 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:

providing a grammar including entries comprising a parent word and a pseudo word being substantially phonetically equivalent to said parent word, said grammar providing a translation from said pseudo word to said parent word;

receiving said parent word as speech;

comparing said speech to said grammar entries; and

selecting said pseudo word as a match for said speech and translating said pseudo word to said parent word.

23. 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:

providing a spelling grammar including one or more entries comprising a parent letter and a pseudo word being substantially phonetically equivalent to said parent letter, said spelling grammar providing a translation from said pseudo word to said parent letter;

receiving said parent letter as speech;

comparing said speech to said grammar entries; and

selecting said pseudo word as a match for said speech and translating said pseudo word to said parent letter.