tailieunhanh - Báo cáo khoa học: "Semantic Information and Derivation Rules for Robust Dialogue Act Detection in a Spoken Dialogue System"

In this study, a novel approach to robust dialogue act detection for error-prone speech recognition in a spoken dialogue system is proposed. First, partial sentence trees are proposed to represent a speech recognition output sentence. Semantic information and the derivation rules of the partial sentence trees are extracted and used to model the relationship between the dialogue acts and the derivation rules. | Semantic Information and Derivation Rules for Robust Dialogue Act Detection in a Spoken Dialogue System Wei-Bin Liang1 Chung-Hsien Wu2 Department of Computer Science and Information Engineering National Cheng Kung University Tainan Taiwan 1 liangnet@ 2chunghsienwu@ Chia-Ping Chen Department of Computer Science and Engineering National Sun Yat-sen University Kaohsiung Taiwan cpchen@ Abstract In this study a novel approach to robust dialogue act detection for error-prone speech recognition in a spoken dialogue system is proposed. First partial sentence trees are proposed to represent a speech recognition output sentence. Semantic information and the derivation rules of the partial sentence trees are extracted and used to model the relationship between the dialogue acts and the derivation 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 evaluated in a Mandarin spoken dialogue system for tour-guiding service. Combined with scores derived from the ASR recognition probability and the dialogue history the proposed approach achieves detection accuracy an absolute improvement of over the baseline of the semantic slot-based method with detection accuracy. 1 Introduction An intuitive framework for spoken dialogue system SDS can be regarded as a chain process. Specifically the automatic speech recognition ASR module accepts the user s utterance Ut and returns a string of words Wt The spoken language understanding SLU module converts Wt to an abstract representation of the user s dialogue act DA . The dialogue management DM module determines the user s dialogue act Aị and accordingly decides the current act of the system. The system DA is converted to a surface representation by natural lan-603 Figure 1 Details of the SLU and DM modules. guage generation in the textual form which is passed to a .