Back to results
US 2006/0190258 A1Published application
N-Best list rescoring in speech recognition
AssigneeNuance Communications, Inc.
InventorsJan Verhasselt, Helmut Dercks
Life of this patent
Estimating the term…
Abstract
A method of speech recognition processing is described based on an N-best list of recognition hypotheses corresponding to a spoken input. Each hypothesis on the N-best list is rescored based on its rank in the rescored N-best list. The rescoring may be based on a Statistical Language Model (SLM) or Dynamic Semantic Model (DSM). One or more rescoring categories may be associated with each recognition hypotheses to affect or bias the rescoring.
Claims19
A method of speech recognition processing comprising:
- producing an N-best list of recognition hypotheses corresponding to a spoken input;
- and
- rescoring the hypotheses to produce a rescored N-best list;
- wherein for each hypothesis the rescoring depends on the rank of the hypothesis in the rescored N-best list.
1 more element
Dependencies are inferred from the claim text.