Paper: Semantic Information and Derivation Rules for Robust Dialogue Act Detection in a Spoken Dialogue System

ACL ID P11-2106
Title Semantic Information and Derivation Rules for Robust Dialogue Act Detection in a Spoken Dialogue System
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2011
Authors

In this study, a novel approach to robust di- alogue act detection for error-prone speech recognition in a spoken dialogue system is proposed. First, partial sentence trees are pro- posed to represent a speech recognition out- put sentence. Semantic information and the derivation rules of the partial sentence trees are extracted and used to model the relation- ship between the dialogue acts and the deriva- tion rules. The constructed model is then used to generate a semantic score for dialogue act detection given an input speech utterance. The proposed approach is implemented and evalu- ated in a Mandarin spoken dialogue system for tour-guiding service. Combined with scores derived from the ASR recognition probabil- ity and the dialogue history, the proposed ap- proach achieves 84.3% detec...