Paper: Combining Acoustic And Pragmatic Features To Predict Recognition Performance In Spoken Dialogue Systems

ACL ID P04-1044
Title Combining Acoustic And Pragmatic Features To Predict Recognition Performance In Spoken Dialogue Systems
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2004
Authors

We use machine learners trained on a combina- tion of acoustic confidence and pragmatic plausi- bility features computed from dialogue context to predict the accuracy of incoming n-best recogni- tion hypotheses to a spoken dialogue system. Our best results show a 25% weighted f-score improve- ment over a baseline system that implements a “grammar-switching” approach to context-sensitive speech recognition.