Robust Loose Coupling for Speech Recognition and Natural Understanding

摘要

The focus of this thesis proposal is to improve the ability of a computationalsystem to understand spoken utterances in a dialogue with a human. Available computational methods for word recognition do not perform as well on spontaneous speech as we would hope. Even a state of the art recognizer achieves slightly worse than 70% word accuracy on (nearly) spontaneous speech in a conversation about a specific problem. To address the incrementality of spontaneously spoken utterances, I will develop methods for segmenting a given utterance into 'chunks' representing individual thoughts. Given an utterance of spontaneous speech, a tool for automatic prosodic feature extraction will analyze the output of the error-correcting post-processor and the acoustic waveform to generate prosodic cues. These cues will aid a robust parser using a prosody-wise grammar to identify the incremental phrases in the utterance and to provide a syntactic analysis. These components will augment the TRAINs-95 conversational planning assistant. (AN).