PRONUNCIATION ACCURACY IN SPEECH RECOGNITION

• US 20140358533A1
• Filed: 04/14/2014
• Published: 12/04/2014
• Est. Priority Date: 05/30/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Nishimura, Masafumi, Tachibana, Ryuki, Kurata, Gakuto
• Granted Patent: US 9,384,730 B2
• Time in Patent Office: Days
• US Class Current: 704/231
• CPC Class Codes: G10L 15/01 Assessment or evaluation of speech recognition systems
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/187 Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams
• CPC Class Codes: G10L 15/197 Probabilistic grammars, e.g. word n-grams

Abstract

A reading accuracy-improving system includes: a reading conversion unit for retrieving a plurality of candidate word strings from speech recognition results to determine the reading of each candidate word string; a reading score calculating unit for determining the speech recognition score for each of one or more candidate word strings with the same reading to determine a reading score; and a candidate word string selection unit for selecting a candidate to output from the plurality of candidate word strings on the basis of the reading score and speech recognition score corresponding to each candidate word string.

Claims

1. A method for improving reading accuracy in speech recognition using processing by a computer, the method comprising computer-executed steps of:

obtaining a plurality of candidate word strings from speech recognition results;

determining a reading of each of the plurality of candidate word strings;

determining a total value of a speech recognition score for each of one or more candidate word strings with the same reading to determine a reading score; and

selecting a candidate among the plurality of candidate word strings to output on the basis of the reading score and the speech recognition score corresponding to each word string.

2. The method for improving reading accuracy according to claim 1, wherein selecting a candidate includes a computer-executed step of weighting and adding together the speech recognition score and the corresponding reading score for each candidate word string to obtain a new score, and selecting the candidate word string with the highest new score.

3. The method for improving reading accuracy according to claim 1, wherein selecting a candidate includes a computer-executed step of selecting a candidate word string with the highest speech recognition score from among the one or more candidate word strings with the highest reading score.

4. The method for improving reading accuracy according to claim 1, wherein determining a total value includes a computer-executed step of allowing for a partial difference in readings between two or more candidate word strings treated as having the same reading in calculating the reading score.

5. The method for improving reading accuracy according to claim 1, wherein the plurality of candidate word strings in the speech recognition results are integrated with N-best lists from a plurality of speech recognition systems.

6. The method for improving reading accuracy according to claim 1, wherein selecting a candidate includes a computer-executed step of selecting all of a plurality of candidate word strings as candidates to be outputted, and rescoring the candidate word strings on the basis of the speech recognition score of each candidate word string and the corresponding reading score.

7. A reading accuracy-improving computer system, comprising:

a computer having a processor and a computer-readable storage device;

a program embodied on the storage device for execution by the processor, the program having a plurality of program modules, the program modules including;

an obtaining module configured to obtain a plurality of candidate word strings from speech recognition results;

a first determining module configured to determine a reading of each of the plurality of candidate word strings;

a second determining module configured to determine a total value of a speech recognition score for each of one or more candidate word strings with the same reading to determine a reading score; and

a selecting module configured to select a candidate among the plurality of candidate word strings to output on the basis of the reading score and the speech recognition score corresponding to each word string.

8. The system according to claim 7, wherein the selecting module includes a computer-executed step of weighting and adding together the speech recognition score and the corresponding reading score for each candidate word string to obtain a new score, and selecting the candidate word string with the highest new score.

9. The system according to claim 7, wherein the selecting modules includes a computer-executed step of selecting a candidate word string with the highest speech recognition score from among the one or more candidate word strings with the highest reading score.

10. The system according to claim 7, wherein the second determining module includes a computer-executed step of allowing for a partial difference in readings between two or more candidate word strings treated as having the same reading in calculating the reading score.

11. The system according to claim 7, wherein a plurality of candidate word strings in the speech recognition results obtained by the obtaining module are integrated with N-best lists from a plurality of speech recognition systems.

12. The system according to claim 7, wherein the selecting module includes a computer-executed step of selecting all of a plurality of candidate word strings as candidates to be outputted, and rescoring the candidate word strings on the basis of the speech recognition score of each candidate word string and the corresponding reading score.

13. A reading-accuracy improving non-transitory computer program product, comprising a computer-readable storage medium having program code embodied therewith, the program code executable by a processor of a computer to perform a method comprising:

obtaining, by the processor, a plurality of candidate word strings from speech recognition results;

determining, by the processor, a reading of each of the plurality of candidate word strings;

determining, by the processor, a total value of a speech recognition score for each of one or more candidate word strings with the same reading to determine a reading score; and

selecting, by the processor, a candidate among the plurality of candidate word strings to output on the basis of the reading score and the speech recognition score corresponding to each word string.

14. The non-transitory computer program product of claim 13, wherein selecting a candidate includes weighting and adding together, by the processor, the speech recognition score and the corresponding reading score for each candidate word string to obtain a new score, and selecting, by the processor, the candidate word string with the highest new score.

15. The non-transitory computer program product of claim 13, wherein selecting a candidate includes selecting, by the processor, a candidate word string with the highest speech recognition score from among the one or more candidate word strings with the highest reading score.

16. The non-transitory computer program product of claim 13, wherein determining a total value of a speech recognition score includes allowing, by the processor, for a partial difference in readings between two or more candidate word strings treated as having the same reading in calculating the reading score.

17. The non-transitory computer program product of claim 13, wherein a plurality of candidate word strings in the speech recognition results, obtained by the processor, are integrated with N-best lists from a plurality of speech recognition systems.

18. The non-transitory computer program product of claim 18, wherein selecting a candidate includes selecting, by the processor, all of a plurality of candidate word strings as candidates to be outputted, and rescoring the candidate word strings, by the processor, on the basis of the speech recognition score of each candidate word string and the corresponding reading score.