Paper: Bayes Risk-based Dialogue Management for Document Retrieval System with Speech Interface

ACL ID C08-2015
Title Bayes Risk-based Dialogue Management for Document Retrieval System with Speech Interface
Venue International Conference on Computational Linguistics
Session Poster Session
Year 2008
Authors
  • Teruhisa Misu (Kyoto University, Kyoto Japan; National Institute of Information and Communications Technology, Kyoto Japan)
  • Tatsuya Kawahara (Kyoto University, Kyoto Japan)

We propose an efficient dialogue manage- ment for an information navigation sys- tem based on a document knowledge base with a spoken dialogue interface. In order to perform robustly for fragmental speech input and erroneous output of an auto- matic speech recognition (ASR), the sys- tem should selectively use N-best hypothe- ses of ASR and contextual information. The system also has several choices in gen- erating responses or confirmations. In this work, we formulate the optimization of the choices based on a unified criterion: Bayes risk, which is defined based on re- ward for correct information presentation and penalty for redundant turns. We have evaluated this strategy with a spoken di- alogue system which also has question- answering capability. Effectiveness of the proposed framew...