System and method for increasing recognition rates of in-vocabulary words by improving pronunciation modeling
View Patent ↗The present disclosure relates to systems, methods, and computer-readable media for generating a lexicon for use with speech recognition. The method includes overgenerating potential pronunciations based on symbolic input, identifying potential pronunciations in a speech recognition context, and storing the identified potential pronunciations in a lexicon. Overgenerating potential pronunciations can include establishing a set of conversion rules for short sequences of letters, converting portions of the symbolic input into a number of possible lexical pronunciation variants based on the set of conversion rules, modeling the possible lexical pronunciation variants in one of a weighted network and a list of phoneme lists, and iteratively retraining the set of conversion rules based on improved pronunciations. Symbolic input can include multiple examples of a same spoken word. Speech data can be labeled explicitly or implicitly and can include words as text and recorded audio.
1. A method comprising:
overgenerating potential pronunciations by converting portions of symbolic input into a number of possible lexical pronunciation variants based on an established set of conversion rules, wherein the symbolic input comprises labeled speech data;
identifying, via a processor of a computing device, potential pronunciations in a speech recognition context to yield identified potential pronunciations; and
storing the identified potential pronunciations in a lexicon.
2. The method of claim 1 , wherein the symbolic input comprises multiple instances of a same spoken word.
3. The method of claim 1 , wherein the labeled speech data is explicitly labeled.
4. The method of claim 1 , wherein the labeled speech data is implicitly labeled.
5. The method of claim 1 , wherein the labeled speech data comprises text and recorded audio.
6. The method of claim 1 , wherein the possible lexical pronunciation variants are modeled in one of a weighted network and phoneme lists.
7. The method of claim 1 , further comprising iteratively retraining the established set of conversion rules based on improved pronunciations.
8. The method of claim 1 , wherein identifying potential pronunciations is based on a threshold.
9. A system comprising:
a processor;
a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
overgenerating potential pronunciations based on symbolic input by converting portions of the symbolic input into a number of possible lexical pronunciation variants based on an established set of conversion rules, wherein the symbolic input comprises labeled speech data;
identifying potential pronunciations in a speech recognition context to yield identified potential pronunciations; and
storing the identified potential pronunciations in a lexicon.
10. The system of claim 9 , wherein the symbolic input comprises multiple instances of a same spoken word.
11. The system of claim 9 , wherein the labeled speech data is one of explicitly labeled and implicitly labeled.
12. The system of claim 9 , wherein the labeled speech data comprises text and recorded audio.
13. The system of claim 9 , wherein the possible lexical pronunciation variants are modeled in one of a weighted network and phoneme lists.
14. The system of claim 9 , the computer-readable storage medium having additional instructions which result in operations comprising iteratively retraining the established set of conversion rules based on improved pronunciations.
15. The system of claim 9 , wherein identifying potential pronunciations is based on a threshold.
16. A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
overgenerating potential pronunciations based on symbolic input by converting portions of the symbolic input into a number of possible lexical pronunciation variants based on an established set of conversion rules, wherein the symbolic input comprises labeled speech data;
identifying potential pronunciations in a speech recognition context to yield identified potential pronunciations; and
storing the identified potential pronunciations in a lexicon.
17. The computer-readable storage device of claim 16 , wherein the labeled speech data comprises text and recorded audio.
18. The computer-readable storage device of claim 16 , wherein the possible lexical pronunciation variants are modeled in one of a weighted network and phoneme lists.
19. The computer-readable storage device of claim 16 , having additional instructions stored which result in operations comprising iteratively retraining the established set of conversion rules based on improved pronunciations.
20. The computer-readable storage device of claim 16 , wherein identifying potential pronunciations is based on a threshold.