Weighting phone confidence measures for automatic speech recognition
A.G.G. Bouwman, Lou Boves, Johan Koolwaaij · Radboud Repository (Radboud University) · 2000
One of the most useful applications of Confidence Measures (CMs) in Automatic Speech Recognition systems is early de tection of incorrect recognition hypotheses. A purely acoustic basis for such a CM is particularly important when tracking errors resulting from Out of Vocabulary speech, background noise or keyword substitution. A commonly taken approach is to compute scores on subword units of the hypothesized words and combine them in a word score. This paper investi gates the assumption that some subword types contain stronger distinctive properties than others. Therefore, their scores ought to have a higher contribution in the eventual word scores. Experiments in a connected digit recognition task showed a relative Confidence Error Rate improvement of 6% on word level and 11% on sentence level in comparison to the baseline CM, with equal contribution of the phone confi dence scores.