An HPM-based Prosodic Analysis of Disfluencies for Spontaneous Mandarin Speech

Cheng Hsien Lin, Chung-Long You, Chen-Yu Chiang, Yih-Ru Wang, Sin-Horng Chen


This paper presents a prosodic analysis of speech disfluencies on MCDC corpus labelled with two types of prosody tags, syllable-juncture break and syllable prosodic state, by a hierarchical prosodic modeling (HPM) approach proposed previously. The prosodic properties of two major types of speech disfluency, repetition and repair, are explored via examining the prosodic tags labelled on all MCDC utterances. The prosodic phrase structure of a disfluency is analyzed by examining the contextual break types. The prosodic state patterns of prosodic phrases in disfluency are also examined. Lastly, the relation between prosodic properties of disfluencies and their pragmatic functions are discussed.


 DOI: 10.21437/SpeechProsody.2018-127

Cite as: Lin, C.H., You, C., Chiang, C., Wang, Y., Chen, S. (2018) An HPM-based Prosodic Analysis of Disfluencies for Spontaneous Mandarin Speech. Proc. 9th International Conference on Speech Prosody 2018, 626-630, DOI: 10.21437/SpeechProsody.2018-127.


@inproceedings{Lin2018,
  author={Cheng Hsien Lin and Chung-Long You and Chen-Yu Chiang and Yih-Ru Wang and Sin-Horng Chen},
  title={An HPM-based Prosodic Analysis of Disfluencies for Spontaneous Mandarin Speech},
  year=2018,
  booktitle={Proc. 9th International Conference on Speech Prosody 2018},
  pages={626--630},
  doi={10.21437/SpeechProsody.2018-127},
  url={http://dx.doi.org/10.21437/SpeechProsody.2018-127}
}