SPEECH RECOGNITION SYSTEM USING MACHINE LEARNING TO CLASSIFY PHONE POSTERIOR CONTEXT INFORMATION AND ESTIMATE BOUNDARIES IN SPEECH FROM COMBINED BOUNDARY POSTERIORS

摘要

A speech recognition system includes a phone classifier and a boundary classifier. The phone classifier generates combined boundary posteriors from a combination of auditory attention features and phone posteriors by feeding phone posteriors of neighboring frames of an audio signal into a machine learning algorithm to classify phone posterior context information. The boundary classifier estimates boundaries in speech contained in the audio signal from the combined boundary posteriors.