Edinburgh Research Explorer

Prosodic Information in a Speech Recognition System intended for Dialogue

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

  • Stephen Isard
  • Simon King
  • Paul Taylor
  • Jacqueline Kowtko

Related Edinburgh Organisations

Original languageEnglish
Title of host publicationIEEE Workshop in speech recognition
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Publication statusPublished - 1995

Abstract

We report on an automatic speech recognition system intended for use in dialogue, whose original aspect is its use of prosodic information for two different purposes. The first is to improve the word level accuracy of the system. The second is to constrain the language model applied to a given utterance by taking into account the way that dialogue context and intonational tune interact to limit the possibilities for what an utterance might be.

ID: 2074665