Dialogue act detection in error-prone spoken dialogue systems using partial sentence tree and latent dialogue act matrix

Wei Bin Liang, Chung-Hsien Wu, Yu Cheng Hsiao

Research output: Contribution to conferencePaper

1 Citation (Scopus)

Abstract

In a goal-oriented spoken dialogue system, the major aim of spoken language understanding is to detect the dialogue acts (DAs) embedded in a speaker's utterance. However, error-prone speech recognition often degrades the performance of the SLU component. In this work, a DA detection approach using partial sentence trees (PSTs) and a latent dialogue act matrix (LDAM) is presented for spoken language understanding. For each input utterance with speech recognition errors, several partial sentences derived from the recognized sentence can be obtained to construct a PST. A set of sentence grammar rules (GRs) is obtained for each partial sentence using the Stanford parser. The relationship between the GRs and the DAs is modeled by an LDAM. Finally, the DA with the highest probability estimated from the speech recognition likelihood, the LDAM and the historical information is determined as the detected DA. In evaluation, compared to the semantic slot-based method which achieved 48.1% dialogue act detection accuracy, the proposed approach can achieve 84.3% accuracy, with 35.2% improvement in accuracy.

Original languageEnglish
Pages3038-3041
Number of pages4
Publication statusPublished - 2010 Dec 1
Event11th Annual Conference of the International Speech Communication Association: Spoken Language Processing for All, INTERSPEECH 2010 - Makuhari, Chiba, Japan
Duration: 2010 Sep 262010 Sep 30

Other

Other11th Annual Conference of the International Speech Communication Association: Spoken Language Processing for All, INTERSPEECH 2010
CountryJapan
CityMakuhari, Chiba
Period10-09-2610-09-30

All Science Journal Classification (ASJC) codes

  • Language and Linguistics
  • Speech and Hearing

Fingerprint Dive into the research topics of 'Dialogue act detection in error-prone spoken dialogue systems using partial sentence tree and latent dialogue act matrix'. Together they form a unique fingerprint.

  • Cite this

    Liang, W. B., Wu, C-H., & Hsiao, Y. C. (2010). Dialogue act detection in error-prone spoken dialogue systems using partial sentence tree and latent dialogue act matrix. 3038-3041. Paper presented at 11th Annual Conference of the International Speech Communication Association: Spoken Language Processing for All, INTERSPEECH 2010, Makuhari, Chiba, Japan.