Multiple Hypotheses in Confidence Scoring Measures for Spoken Dialog System

摘要

This study provides improved confidence assessment for detection of word-level speech recognition errors, out of domain utterances and incorrect concepts. New features from the speech understanding component are proposed for confidence annotation at utterance and concept levels. In addition, the confidence measures are used to improve the word recognition accuracy. Several hypotheses from different speech recognizers are compiled into a word-graph. The word-graph is searched for the hypothesis with the best confidence.