5th European Conference on Speech Communication and Technology

Rhodes, Greece
September 22-25, 1997

Using Intonation to Constrain Language Models in Speech Recognition

Paul Taylor, Simon King, Stephen Isard, Helen Wright, Jacqueline Kowtko

Centre for Speech Technology Research, University of Edinburgh, UK

This paper describes a method for using intonation to reduce word error rate in a speech recognition system designed to recognise spontaneous dialogue speech. We use a form of dialogue analysis based on the theory of conversational games. Different move types under this analysis conform to different language models. Different move types are also characterised by different into-national tunes. Our overall recognition strategy is first to predict from intonation the type of game move that a test utterance represents, and then to use a bigram language model for that type of move during recognition.

Full Paper

Bibliographic reference.  Taylor, Paul / King, Simon / Isard, Stephen / Wright, Helen / Kowtko, Jacqueline (1997): "Using intonation to constrain language models in speech recognition", In EUROSPEECH-1997, 2763-2766.