METHODS AND APPARATUS FOR JOINT STOCHASTIC AND DETERMINISTIC DICTATION FORMATTING

摘要

Methods and apparatus for speech recognition on user dictated words to generate a dictation and using a discriminative statistical model derived from a deterministic formatting grammar module and user formatted documents to extract features and estimate scores from the formatting graph. The processed dictation can be output as formatted text based on a formatting selection to provide an integrated stochastic and deterministic formatting of the dictation.

主权项

1. A method, comprising:
performing, using a computer processor, speech recognition on user dictated words to generate a dictation; parsing the dictation using a deterministic formatting grammar module to build a formatting graph; using a discriminative statistical model derived from a deterministic formatting grammar module and user formatted documents to extract features from the formatting graph and estimate scores; choosing a path in the formatting graph as a formatting selection; and outputting the dictation as formatted text based on the formatting selection to provide an integrated stochastic and deterministic formatting of the dictation for disambiguation of the user dictated words.