Workshop Homepage
Workshop 2000
TEAM GOALS
Project Description
Final Report (in .pdf and .doc formats)


TEAM MEMBERS
The Team

  Name E-Mail Affiliation
Fung, Pascale Pascale@ee.ust.hk HKUST
Byrne, William byrne@jhu.edu CLSP/JHU
Zheng, Tom fzheng@sp.cs.tsinghua.edu.cn Tsinghua University
Venkatramani, Veera veera@jhu.edu CLSP/JHU
Ruhi, Umar G7umar@cdf.toronto.edu University of Toronto
Yi, Liu eelyx@ee.ust.hk HKUST
Song, Zhanjiang szj@sp.cs.tsinghua.edu.cn Tsinghua University
Kamm, Terri tkamm@clsp.jhu.edu Department of Defense

WORKSHOP PRESENTATIONS
Opening Day Presentation (7/17)  

CASS Speech Corpus -- Li Aijun, Asociate Professor, Chinese Academy of Social Sciences (7/19)

First Research Update from the MCS group (7/26)

Final Presentation from the MCS group (8/23)

GROUP PUBLICATIONS
  • CASS: A Phonetically Transcribed Corpus Of Mandarin Spontaneous Speech , (ps) (pdf)
    LI Aijun, ZHENG Fang, William Byrne, Pascale Fung, Terri Kamm, LIU Yi, SONG Zhanjiang, Umar Ruhi, Veera Venkataramani, CHEN XiaoXia
    International Conference on Spoken Language Processing, October, 2000, Beijing, China
  • Modelling Pronunciation Variations in Spontaneous Mandarin speech, (ps) (pdf)
    Liu Yi and Pascale Fung
    International Conference on Spoken Language Processing, October, 2000, Beijing, China
  • Mandarin pronunciation variation modeling,
    F. Zheng, Z. Song, P. Fung and W.Byrne,
    Proceedings of NCMMSC6, ShenZhen, China, 2001.
  • Reducing pronunciation lexicon confusion and using more data without phonetic transcription for pronunciation modeling,
    F. Zheng, Z. Song, P. Fung and W.Byrne,
    Proceedings of the 7th International Conference on Spoken Language Processing, Denver, Colorado, September 2002.
  • Modeling pronunciation variation using context-dependent weighting and B/S refined acoustic modeling,
    F. Zheng, Z.J. Song, P. Fung and W. Byrne,
    Proceedings of the European Conference on Speech Communication and Technology (Eurospeech), Alborg, Denmark, pp. 57-60, 2001.
  • Pronunciation Modeling of Spontaneous Mandarin Speech Using Phonetic Feature Distance and Optimal Gaussian Mixture Sharing,
    Liu Yi and Pascale Fung, William Byrne and Urma Ruhi,
    student forum, Proceedings of the IEEE International Conference on Acoustic Speech and Signal Processing, Salt Lake City, USA, 2001.
  • Estimating Pronunciation Variations from Acoustic Likelihood Score for HMM Reconstruction,
    Liu Yi and Pascale Fung,
    Proceedings of the European Conference on Speech Communication and Technology, Alborg, Denmark, 2001.
  • Partial Change Phone Models for Pronunciation Variations in Spontaneous Mandarin Speech,
    Liu Yi and Pascale Fung,
    Proceedings of the International Symposium on Chinese Spoken Language Processing, Taipei, Taiwan, 2002.
  • Model Partial Pronunciation Variations for Spontaneous Mandarin Speech Recognition,
    Liu Yi and Pascale Fung,
    Proceedings of the 7th International Conference on Spoken Language Processing, Denver, Colorado, September 2002.
  • Pronunciation Modeling for Spontaneous Mandarin Speech Recognition,
    Liu Yi and Pascale Fung,
    Submitted to the International Journal of Speech Technology.
  • Triphone Model Reconstruction for Mandarin Variations,
    Liu Yi and Pascale Fung,
    will appear in Proceedings of the IEEE International Conference on Acoustic Speech and Signal Processing, HongKong, April 2003.
  • Modeling Partial Pronunciation Variations for Spontaneous Mandarin Speech Recognition,
    Liu Yi and Pascale Fung,
    to appear in Computer Speech and Language.

    Final Team Technical Reports

    THE PAPER CLUB

    WORKSHOP TOOLS
    During the workshop, we wrote some generic alignment tools. These were built on top of the AT&T fsmtoolkit. You can download them here.