Using Prosodic Information to Constrain Language Models for Spoken dialogue

Paul Taylor, Hiroshi Shimodaira, Stephen Isard, Simon King, Jacqueline Kowtko

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

We present work intended to improve speech recognition performance for computer dialogue by taking into account the way that dialogue context and intonational tune interact to limit the possibilities for what an utterance might be. We report here on the extra constraint achieved in a bigram language model expressed in terms of entropy by using separate submodels for different sorts of dialogue acts and trying to predict which submodel to apply by analysis of the intonation of the sentence being recognised.
Original languageEnglish
Title of host publicationProceedings of the Fourth International Conference on Spoken Language, 1996 (ICSLP `96)
Number of pages4
Publication statusPublished - Oct 1996


Dive into the research topics of 'Using Prosodic Information to Constrain Language Models for Spoken dialogue'. Together they form a unique fingerprint.

Cite this