Classification of dialogue acts using prosodic features in chinese spontaneous speech

被引:0
|
作者
Sun, H [1 ]
Xu, MX [1 ]
Wu, WH [1 ]
机构
[1] Tsinghua Univ, State Key Lab Intelligent Technol & Syst, Ctr Speech Technol, Beijing 100084, Peoples R China
关键词
dialogue acts (DAs); speech understanding; prosody; spoken dialogue system; spontaneous speech;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Dialogue acts (DAs) indicate the meanings of the utterances at the aspect of speech act, e.g. whether an utterance is a question, statement, confirmation and so forth, which helps to understand the utterances. The understanding component of a dialogue system can be improved by knowing the dialogue acts of utterances. In this paper, we investigate the classification of the dialogue acts using prosodic features in EasyFlight, a Chinese spoken dialogue system for querying and booking flight tickets. The experiments are based on the domain-specified Chinese spontaneous speech corpus, which is collected for evaluating the performance of the dialogue system. Dialogue Acts are labeled by hand. According to more than thirty prosodic features two types of DAs, Question and Statement are classified in the experiments, and the classification accuracy of 83% is obtained.
引用
收藏
页码:1163 / 1166
页数:4
相关论文
共 50 条
  • [1] Automatic classification of question turns in spontaneous speech using lexical and prosodic evidence
    Ananthakrishnan, Sankaranarayanan
    Ghosh, Prasanta
    Narayanan, Shrikanth
    [J]. 2008 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-12, 2008, : 5005 - 5008
  • [2] Using prosodic cues to identify dialogue acts: methodological challenge
    Sikveland, Rein Ove
    Zeitlyn, David
    [J]. TEXT & TALK, 2017, 37 (03) : 311 - 334
  • [3] ACQUISITION OF PROSODIC FEATURES OF SPEECH USING A SPEECH VISUALIZER
    JAMES, EF
    [J]. IRAL-INTERNATIONAL REVIEW OF APPLIED LINGUISTICS IN LANGUAGE TEACHING, 1976, 14 (03): : 227 - 243
  • [4] SPEECH EMOTION CLASSIFICATION USING SVM AND MLP ON PROSODIC AND VOICE QUALITY FEATURES
    Idris, Inshirah
    Salam, Md Sah Hj
    Sunar, Mohd Shahrizal
    [J]. JURNAL TEKNOLOGI, 2016, 78 (2-2): : 27 - 33
  • [5] Synthesizing dialogue speech of Japanese based on the quantitative analysis of prosodic features
    Hirose, K
    Sakata, M
    Kawanami, H
    [J]. ICSLP 96 - FOURTH INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE PROCESSING, PROCEEDINGS, VOLS 1-4, 1996, : 378 - 381
  • [6] Dravidian language classification from speech signal using spectral and prosodic features
    Koolagudi S.G.
    Bharadwaj A.
    Srinivasa Murthy Y.V.
    Reddy N.
    Rao P.
    [J]. International Journal of Speech Technology, 2017, 20 (4) : 1005 - 1016
  • [7] Automatically predicting dialogue structure using prosodic features
    Hastie, HW
    Poesio, M
    Isard, S
    [J]. SPEECH COMMUNICATION, 2002, 36 (1-2) : 63 - 79
  • [8] EVALUATION OF MIMICKED SPEECH USING PROSODIC FEATURES
    Mary, Leena
    Babu, Anish K. K.
    Joseph, Aju
    George, Gibin M.
    [J]. 2013 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2013, : 7189 - 7193
  • [9] Predicting dialogue acts from prosodic information
    Olguin, SRC
    Cortés, LAP
    [J]. COMPUTATIONAL LINGUISTICS AND INTELLIGENT TEXT PROCESSING, 2006, 3878 : 355 - 365
  • [10] Spectral and prosodic features-based speech pattern classification
    Sinha, Shweta
    Jain, Aruna
    Agrawal, S. S.
    [J]. INTERNATIONAL JOURNAL OF APPLIED PATTERN RECOGNITION, 2015, 2 (01) : 96 - 110