Method and apparatus for improving the transcription accuracy of speech recognition software
First Claim
1. A method for operating a computerized speech recognition system using a computer-based system, said computer-based system performing the steps of:
- loading a first vocabulary, said first vocabulary containing inputs provided from the inputs of previous users of the computer-based system;
evaluating individual vocabulary elements within said first vocabulary to determine a first vocabulary match set, said vocabulary elements within said first vocabulary match set having match probability scores;
weighting said match probability scores of said vocabulary elements within said first vocabulary match set with a first vocabulary weighting factor;
loading a second vocabulary, said second vocabulary containing inputs provided from the inputs of previous users of the computer-based system;
evaluating individual vocabulary elements within said second vocabulary to determine a second vocabulary match set, said vocabulary elements within said second vocabulary match set having match probability scores;
combining with said computer-based system said individual vocabulary elements within said first and second vocabulary match sets so as to create a combined set of vocabulary elements, said combined set of vocabulary elements including weighted match probability scores of said vocabulary elements of said first vocabulary match set and unweighted match probability scores of said vocabulary elements of said second vocabulary match set;
weighting said match probability scores of said combined set of vocabulary elements with a second vocabulary weighting factor; and
selecting as a match to an input to said computerized speech recognition system a vocabulary element from said combined set of vocabulary elements based on said weighted match probability scores of said combined set of vocabulary elements.
0 Assignments
0 Petitions
Accused Products
Abstract
The present invention involves an enhanced method for operating a speech recognition system in which sequential vocabularies are loaded for comparison to the input speech. Form each sequential vocabulary a subset of candidate vocabulary elements are selected for matching, the probability match scores of each of the selected candidate vocabulary elements is weighted by a weighting factor that is dependant on the particular vocabulary from which the vocabulary elements are selected. As each set of candidate vocabulary elements is selected from the next sequential vocabulary, the selected set is combined with the set of previously selected and weighted vocabulary elements, which is then further weighted according to the weighting criteria of the then active vocabulary. At the end of the sequential vocabulary selection grouping process, the final, sequentially weighted match scores for the candidate vocabulary elements are arranged and an appropriate match to the input speech is presented.
279 Citations
14 Claims
-
1. A method for operating a computerized speech recognition system using a computer-based system, said computer-based system performing the steps of:
-
loading a first vocabulary, said first vocabulary containing inputs provided from the inputs of previous users of the computer-based system; evaluating individual vocabulary elements within said first vocabulary to determine a first vocabulary match set, said vocabulary elements within said first vocabulary match set having match probability scores; weighting said match probability scores of said vocabulary elements within said first vocabulary match set with a first vocabulary weighting factor; loading a second vocabulary, said second vocabulary containing inputs provided from the inputs of previous users of the computer-based system; evaluating individual vocabulary elements within said second vocabulary to determine a second vocabulary match set, said vocabulary elements within said second vocabulary match set having match probability scores; combining with said computer-based system said individual vocabulary elements within said first and second vocabulary match sets so as to create a combined set of vocabulary elements, said combined set of vocabulary elements including weighted match probability scores of said vocabulary elements of said first vocabulary match set and unweighted match probability scores of said vocabulary elements of said second vocabulary match set; weighting said match probability scores of said combined set of vocabulary elements with a second vocabulary weighting factor; and selecting as a match to an input to said computerized speech recognition system a vocabulary element from said combined set of vocabulary elements based on said weighted match probability scores of said combined set of vocabulary elements. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A method for operating a computerized speech recognition system using a computer-based system, said computer-based system performing the steps of:
-
loading a first vocabulary, said first vocabulary containing inputs provided from the inputs of previous users of the computer-based system; evaluating individual vocabulary elements within said first vocabulary to determine a first vocabulary match set, said vocabulary elements within said first vocabulary match set having match probability scores; loading a second vocabulary, said second vocabulary containing inputs provided from the inputs of previous users of the computer-based system; evaluating individual vocabulary elements within said second vocabulary to determine a second vocabulary match set, said vocabulary elements within said second vocabulary match set having match probability scores; weighting said match probability scores of said vocabulary elements within said first vocabulary match set with a prioritization factor; combining with said computer-based system said individual vocabulary elements within said first and second vocabulary match sets so as to create a combined set of vocabulary elements, said combined set of vocabulary elements including weighted match probability scores of said vocabulary elements of said first vocabulary match set and unweighted match probability scores of said vocabulary elements of said second vocabulary match set; normalizing said match probability scores of said combined set of vocabulary elements; weighting said match probability scores of said normalized combined set of vocabulary elements with a non-linear weighting function; and selecting as a match to an input to said computerized speech recognition system a vocabulary element from said combined set of vocabulary elements based on said non-linearly weighted match probability scores of said combined set of vocabulary elements. - View Dependent Claims (10, 11, 12, 13)
-
-
14. A non-transitory computer-readable media having executable instructions for causing a processor to perform a method for operating a computerized, speech recognition system, said method comprising:
-
loading a first vocabulary, said first vocabulary containing inputs provided from the inputs of previous users of the computer-based system; evaluating individual vocabulary elements within said first vocabulary to determine a first vocabulary match set, said vocabulary elements within said first vocabulary match set having match probability scores; weighting said match probability scores of said vocabulary elements within said first vocabulary match set with a first vocabulary weighting factor; loading a second vocabulary, said second vocabulary containing inputs provided from the inputs of previous users of the computer-based system; evaluating individual vocabulary elements within said second vocabulary to determine a second vocabulary match set, said vocabulary elements within said second vocabulary match set having match probability scores; combining with said computer-based system said individual vocabulary elements within said first and second vocabulary match sets so as to created a combined set of vocabulary elements, said combined set of vocabulary elements including weighted match probability scores of said vocabulary elements of said first vocabulary match set and unweighted match probability scores of said vocabulary elements of said second vocabulary match set, weighting said match probability scores of said combined set of vocabulary elements with a second vocabulary weighting factor; and selecting as a match to an input to said computerized speech recognition system a vocabulary element from said combined set of vocabulary elements based on said weighted match probability scores of said combined set of vocabulary elements.
-
Specification