System and Method of Automated Language Model Adaptation

摘要

Systems and methods of automated adaptation of a language model for transcription of audio data include obtaining audio data. The audio data is transcribed with a language model to produce a plurality of audio file transcriptions. A quality of the plurality of audio file transcriptions is evaluated. At least one best transcription from a plurality of audio file transcriptions is selected based upon the evaluated quality. Statistics are calculated from the selected at least one best transcription from the plurality of audio file transcriptions. The language model is modified from the calculated statistics.

主权项

1. A method of automated adaptation of a language model for transcription of audio data, the method comprising:
obtaining audio data from a plurality of audio files; transcribing the audio data with a language model to produce a plurality of audio file transcriptions, at least one audio file transcription of the plurality being associated to each audio file of the plurality of audio files; evaluating a quality of the plurality of audio file transcriptions; selecting at least one best transcription from the plurality of audio file transcriptions based upon the evaluated quality; calculating statistics from the selected at least one best transcription from the plurality of audio file transcriptions; and modifying the language model from the calculated statistics.