Speech-recognition grammar analysis
View Patent ↗An apparatus for determining at least one recognition parameter, associated with a grammar, for use in attempting to recognize utterances using the grammar includes an input configured to receive indicia of the grammar, an output configured to provide the at least one recognition parameter, and a processor coupled to the input and to the output and configured to analyze words of the grammar to determine a feature of the grammar and to provide indicia of the at least one recognition parameter to the output based on the determined feature.
1. An apparatus comprising:
an input that receives indicia of a grammar;
an output that provides a recognition parameter associated with said grammar;
a processor coupled to said input and to said output and
a computer-readable medium including computer executable instructions that, when executed by said processor, cause said processor to:
analyze said grammar to determine a perplexity and word length associated with said grammar;
determine a category of said grammar based on said perplexity and word length;
determine said recognition parameter based on said category; and
provide indicia of said recognition parameter to said output.
2. The apparatus of claim 1 , wherein said recognition parameter is selected from the group consisting of a confidence level, a pruning value, a noise reduction, a word-boundary acoustic evaluation, a word-internal acoustic evaluation, a model selection, and a timeout value.
3. The apparatus of claim 1 , wherein said input is coupled to a packet-switched communication network, and said output communicates with a speech recognizer.
4. A non-transitory medium comprising computer-readable instructions that, when read by a computer, cause said computer to:
receive indicia of a grammar;
analyze said grammar to determine a perplexity and word length associated with said grammar;
determine a category of said grammar based on said perplexity and word length;
determine a recognition parameter based on said category; and
output indicia of said recognition parameter based on said category.
5. The non-transitory medium of claim 4 , wherein said recognition parameter is selected from the group consisting of a confidence level, a pruning value, a noise reduction, a word-boundary acoustic evaluation, a word-internal acoustic evaluation, a model selection, and a timeout value.
6. A system comprising:
an input that receives an utterance;
a first module that:
analyzes grammar to determine a perplexity and word length associated with said grammar;
determines a category of said grammar based on said perplexity and word length; and
determines a recognition parameter based on said category; and
a second module, coupled to said first module and said input, that:
applies a speech model in accordance with said recognition parameter in an attempt to recognize said utterance.
7. The system of claim 6 , wherein said recognition parameter is selected from the group consisting of a confidence level, a pruning value, a noise reduction, a word-boundary acoustic evaluation, a word-internal acoustic evaluation, a model selection, and a timeout value.
8. The system of claim 6 , wherein said input is coupled to a packet-switched communication network.