PRONUNCIATION ACCURACY IN SPEECH RECOGNITION

• US 20160210964A1
• Filed: 03/28/2016
• Published: 07/21/2016
• Est. Priority Date: 05/30/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kurata, Gakuto, Nishimura, Masafumi, Tachibana, Ryuki
• Granted Patent: US 9,978,364 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• 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:

program instructions to obtain a plurality of candidate word strings from speech recognition results, wherein the speech recognition results contain a speech recognition score for each of the plurality of candidate work strings;

program instructions to determine a reading of each of the plurality of candidate word strings, wherein two or more candidate word strings have the same reading, and wherein the two or more candidate word strings having the same reading are homophones;

program instructions to determine a reading score for each candidate word string, wherein the reading score for each of the two or more candidate word strings with the same reading is based on a total value of the speech recognition scores for the two or more candidate word strings with the same reading; and

program instructions 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, wherein program instructions to select the candidate includes a computer system-executed step selected from the group consisting of;

(a) program instructions to weight and add together the speech recognition score and the corresponding reading score for each candidate word string to obtain a new score, and program instructions to select the candidate word string with the highest new score; and

(b) program instructions to select a candidate word string with the highest speech recognition score from among the one or more candidate word strings with the highest reading score.

2. The method for improving reading accuracy according to claim 1, wherein determining the total value of the speech recognition scores for the two or more candidate word strings with the same reading includes computer-executed step steps of:

determining two or more candidate word strings with partial tolerable different readings to be allowing for a partial difference in readings between two or more candidate word strings treated as having the same reading, partial tolerable different readings being a first predetermined value, the partial difference in readings range being a predetermined range of values; and

calculating the reading score total value of the speech recognition scores for the two or more candidate word strings with the same reading includes speech recognition scores for the two or more candidate word strings with partial tolerable different readings to be treated as having the same reading.

3. The method for improving reading accuracy according to claim 1, wherein the homophones are homophones with different notations.

4. 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.

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

6. A computer program product for generating an index for a table in a database, the computer program product comprising:

one or more computer-readable storage media and program instructions stored on the one or more computer-readable storage media, the program instructions comprising;

program instructions to obtain a plurality of candidate word strings from speech recognition results, wherein the speech recognition results contain a speech recognition score for each of the plurality of candidate work strings;

program instructions to determine a reading of each of the plurality of candidate word strings, wherein two or more candidate word strings have the same reading, and wherein the two or more candidate word strings having the same reading are homophones;

program instructions to determine a reading score for each candidate word string, wherein the reading score for each of the two or more candidate word strings with the same reading is based on a total value of the speech recognition scores for the two or more candidate word strings with the same reading; and

program instructions 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, wherein program instructions to select the candidate includes a computer program product-executed step selected from the group consisting of;

(a) program instructions to weight and add together the speech recognition score and the corresponding reading score for each candidate word string to obtain a new score, and program instructions to select the candidate word string with the highest new score; and

(b) program instructions to select a candidate word string with the highest speech recognition score from among the one or more candidate word strings with the highest reading score.

7. The computer program product according to claim 6, wherein program instructions to determine the total value of the speech recognition scores for the two or more candidate word strings with the same reading includes computer program product-executed step steps of:

program instructions to determine two or more candidate word strings with partial tolerable different readings value to be program instructions to allow for a partial difference in readings range between two or more candidate word strings treated as having the same reading, partial tolerable different readings being a first predetermined value, the partial difference in readings range being a predetermined range of values; and

program instructions to calculate the reading score total value of the speech recognition scores for the two or more candidate word strings with the same reading includes speech recognition scores for the two or more candidate word strings with partial tolerable different readings value to be treated as having the same reading.

8. The computer program product according to claim 6, wherein the homophones are homophones with different notations.

9. The computer program product according to claim 6, 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.

10. The computer program product according to claim 6, wherein program instructions to select a candidate includes the computer program product-executed step of program instructions to select all of a plurality of candidate word strings as candidates to be outputted, and program instructions to re-score the candidate word strings on the basis of the speech recognition score of each candidate word string and the corresponding reading score.

11. A computer system for improving reading accuracy in speech recognition, the computer system comprising:

one or more computer processors;

one or more computer-readable storage media;

program instructions stored on the computer-readable storage media for execution by at least one of the one or more processors, the program instructions comprising;

program instructions to obtain a plurality of candidate word strings from speech recognition results, wherein the speech recognition results contain a speech recognition score for each of the plurality of candidate work strings;

program instructions to determine a reading of each of the plurality of candidate word strings, wherein two or more candidate word strings have the same reading, and wherein the two or more candidate word strings having the same reading are homophones;

program instructions to determine a reading score for each candidate word string, wherein the reading score for each of the two or more candidate word strings with the same reading is based on a total value of the speech recognition scores for the two or more candidate word strings with the same reading; and

program instructions 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, wherein program instructions to select the candidate includes a computer system-executed step selected from the group consisting of;

(a) program instructions to weight and add together the speech recognition score and the corresponding reading score for each candidate word string to obtain a new score, and program instructions to select the candidate word string with the highest new score; and

(b) program instructions to select a candidate word string with the highest speech recognition score from among the one or more candidate word strings with the highest reading score.

12. The system according to claim 11, wherein program instructions to determine the total value of the speech recognition scores for the two or more candidate word strings with the same reading includes computer system-executed step steps of:

program instructions to determine two or more candidate word strings with partial tolerable different readings value to be program instructions to allow for a partial difference in readings range between two or more candidate word strings treated as having the same reading, partial tolerable different readings being a first predetermined value, the partial difference in readings range being a predetermined range of values; and

program instructions to calculate the reading score total value of the speech recognition scores for the two or more candidate word strings with the same reading includes speech recognition scores for the two or more candidate word strings with partial tolerable different readings value to be treated as having the same reading.

13. The system according to claim 11, wherein the homophones are homophones with different notations.

14. The system according to claim 11, 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.

15. The system according to claim 11, wherein program instructions to select a candidate includes the computer system-executed step of program instructions to select all of a plurality of candidate word strings as candidates to be outputted, and program instructions to re-score the candidate word strings on the basis of the speech recognition score of each candidate word string and the corresponding reading score.