Method and system for dynamic creation of contexts
View Patent ↗A method and a system for a speech recognition system, comprising an electronic speech-based document is associated with a document template and comprises one or more sections of text recognized or transcribed from sections of speech. The sections of speech are transcribed by the speech recognition system into corresponding sections of text of the electronic speech based document. The method includes the steps of dynamically creating sub contexts and associating the sub context to sections of text of the document template.
1. A method for converting speech dictated by a user to text in a speech based document, the method comprising:
associating a basic context with the speech based document, the basic context comprising a first lexicon and first language model;
associating a first sub context, of the basic context, with a first portion of the speech based document, the sub context comprising a second lexicon and a second language model;
recognizing speech corresponding to the first portion using the first sub context; and
recognizing speech corresponding to portions of the electronic document to which no sub context has been associated using the basic context.
2. The method of claim 1 , wherein the basic context further comprises a first grammar.
3. The method of claim 2 , wherein the first sub context further comprises a second grammar.
4. The method of claim 1 , further comprising associating a second sub context, of the basic context, with a second portion of the speech based document, the second sub context comprising a third lexicon and a third language model, and recognizing speech corresponding to the second portion using the second sub context.
5. The method of claim 4 , wherein the second sub context further comprises a third grammar.
6. A system for converting speech dictated by a user to text n a speech based document, the system comprising:
at least one processor configured to:
associating a basic context with the speech based document, the basic context comprising a first lexicon and first language model;
associating a first sub context, of the basic context, with a first portion of the speech based document, the sub context comprising a second lexicon and a second language model;
recognizing speech corresponding to the first portion using the first sub context; and
recognizing speech corresponding to portions of the electronic document to which no sub context has been associated using the basic context.
7. The system of claim 6 , wherein the basic context further comprises a first grammar.
8. The system of claim 7 , wherein the first sub context further comprises a second grammar.
9. The system of claim 6 , wherein the at least one processor is configured to:
associate a second sub context with a second portion of the speech based document, the second sub context comprising third lexicon and a third language model; and
recognize speech corresponding to the second portion using the second sub context.
10. The system of claim 9 , wherein the second sub context further comprises a third grammar.