User-aided adaptation of a phonetic dictionary
First Claim
1. A method for adapting a phonetic dictionary for peculiarities of a speech of at least one speaker, the method comprising:
- a. receiving a search term in text form;
b. searching for the search term in the phonetic dictionary, wherein if the search term or a portion thereof is included in the phonetic dictionary,obtaining from the phonetic dictionary pronunciation elements corresponding to phonetic elements of the search term or a portion thereof, for serving as search pronunciations of the search term or portion thereof,and if the search term or portion thereof is not included in the phonetic dictionary,generating pronunciation elements corresponding to phonetic elements of the search term or portion thereof with a text to phoneme model (TTP); and
concatenating all the obtained and generated pronunciation elements of the search term to form a search pronunciation of the search term;
c. searching by the formed search pronunciation for a matching phonetic transcription in an indexed transcriptions database, wherein the indexed transcriptions database stores phonetic transcriptions of a corresponding audio database of recorded human speech;
wherein the searching yields a found pronunciation;
d. retrieving an audio section corresponding to the matching phonetic transcription of the found pronunciation from the audio database;
e. audibly presenting to a person the audio section and tagging the audio section according to its acceptability for pronouncing of the search term to obtain a tagged search result;
f. calculating a value of acceptability efficiency level for the found pronunciation, as obtained from the indexed transcriptions database, wherein the value of acceptability determined by the tagging; and
g. in case the calculated value of acceptability exceeds a certain threshold value and the found pronunciation, as obtained from the indexed transcriptions database, is not already included in the phonetic dictionary, adding the found pronunciation to the phonetic dictionary;
wherein the method is performed on an at least one computerized apparatus configured to perform the method.
3 Assignments
0 Petitions
Accused Products
Abstract
A method for adapting a phonetic dictionary for peculiarities of a speech of an at least one speaker, comprising generating search pronunciations for a search term, retrieving audio sections from an audio database for each search pronunciation, audibly presenting to a person the audio sections of the speech of the at least one speaker, and updating the phonetic dictionary based on acceptability of the audio sections determined from judgments by the person regarding intelligibility of the audio sections in audibly pronouncing the provided at least one word, wherein the method is performed on an at least one computerized apparatus configured to perform the method.
-
Citations
16 Claims
-
1. A method for adapting a phonetic dictionary for peculiarities of a speech of at least one speaker, the method comprising:
-
a. receiving a search term in text form; b. searching for the search term in the phonetic dictionary, wherein if the search term or a portion thereof is included in the phonetic dictionary, obtaining from the phonetic dictionary pronunciation elements corresponding to phonetic elements of the search term or a portion thereof, for serving as search pronunciations of the search term or portion thereof, and if the search term or portion thereof is not included in the phonetic dictionary, generating pronunciation elements corresponding to phonetic elements of the search term or portion thereof with a text to phoneme model (TTP); and concatenating all the obtained and generated pronunciation elements of the search term to form a search pronunciation of the search term; c. searching by the formed search pronunciation for a matching phonetic transcription in an indexed transcriptions database, wherein the indexed transcriptions database stores phonetic transcriptions of a corresponding audio database of recorded human speech;
wherein the searching yields a found pronunciation;d. retrieving an audio section corresponding to the matching phonetic transcription of the found pronunciation from the audio database; e. audibly presenting to a person the audio section and tagging the audio section according to its acceptability for pronouncing of the search term to obtain a tagged search result; f. calculating a value of acceptability efficiency level for the found pronunciation, as obtained from the indexed transcriptions database, wherein the value of acceptability determined by the tagging; and g. in case the calculated value of acceptability exceeds a certain threshold value and the found pronunciation, as obtained from the indexed transcriptions database, is not already included in the phonetic dictionary, adding the found pronunciation to the phonetic dictionary; wherein the method is performed on an at least one computerized apparatus configured to perform the method. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
Specification