Kai Yu

Kai Yu entered the Department of Automation at Tsinghua University, P.R.China in 1994. After obtaining B.Eng. in 1999, he continued his study for M.Sc. in the State Key Laboratory of Pattern Recognition and Intelligence Systems in the same department untill 2002. He then joined the Machine Intelligence Laboratory (MIL) (formerly the Speech Vision and Robotics (SVR) group) for PhD under the supervision of Dr Mark Gales. He obtained his PhD in 2006 and is currently working as a senior research associate in MIL. Other academic staffs in the speech group are Prof. Steve Young, Prof. Phil. Woodland and Dr. Bill Byrne. He was a member of Hughes Hall college in Cambridge University.

Research interests

• Adaptation and adaptive training in speech recognition and synthesis
• Discriminative training of acoustic models
• HMM based statistical speech synthesis
• Spoken dialogue system, speech understanding and human computer interaction
• Bayesian inference and graphical model in pattern recognition
• Statistical learning theory and kernel methods
• Signal processing, image processing and understanding
• Application of machine intelligence
• Spoken language technology for language learning

Current projects

• Automatic Communication Skill Evaluation. (to start)
• Probabalistic Adaptive Learning And Natural Conversational Engine (PARLANCE) (EU funded project: to start in Nov. 2011) (Research Associate position available)
• Robust dialogue for infotainment. (Research Associate position available)

Past projects

• Computational Learning in Adaptive Systems for Spoken Conversation (CLASSIC) (EU funded project)
• Global Autonomous Language Exploitation (GALE) (DARPA funded project)
• Effective, Affordable Reusable Speech-to-text (EARS) (DARPA funded project)

Activities

• Area chair (speech processing) and technical committee member of EUSIPCO 2011
• Publication chair of IEEE ASRU 2011
• Programme committee member of IUI 2010
• Area chair (speech recognition) and technical committee member of INTERSPEECH 2009
• Regular reviewer for internatiaonal conferences and journals
  Conferences: ICASSP, INTERSPEECH, ASRU, EACL, IUI, CHI, EUSIPCO
  Journals: IEEE Trans. on ASL, IEEE Signal Processing Letters, Speech Communication, Computer Speech & Language, IET Signal Processing Letters
• Research grant reviewer
  National Natural Science Foundation of China (NSFC)
  Foundation for Polish Science (FNP)

Publications

Thesis

Journal Papers

• K. Yu, H. Zen, F. Mairesse and S. Young. (2011)
  Context adaptive training with factorized decision trees for HMM-based statistical parametric speech synthesis.
  Speech Communication, vol.53, no.6, 914--923, 2011.
• K. Yu and S. Young (2011).
  Continuous F0 modelling for HMM based statistical parametric speech synthesis.
  IEEE Transactions on Audio, Speech and Language Processing, vol.19, no.5, 1071--1079, 2011.
• K. Yu, M. J. F. Gales, L. Wang and P. C. Woodland (2010).
  Unsupervised training and directed manual transcription for LVCSR.
  Speech Communication, 52, 652--663, 2010.
  
• S. Young, M. Gasic, S. Keizer, F. Mairesse, J. Schatzmann, B. Thomson and K. Yu (2009).
  The hidden information state model: a practical framework for POMDP-based spoken language management.
  Computer Speech and Language, vol. 24, no. 2, 150--174, 2009.
• K. Yu, M. J. F. Gales and P. C. Woodland (2009).
  Unsupervised adaptation with discriminative mapping transforms.
  IEEE Transactions on Audio, Speech and Language Processing, vol.17, no.4, 714--723, 2009.
• K. Yu and M. J. F. Gales (2007).
  Bayesian adaptive inference and adaptive training.
  IEEE Transactions on Audio, Speech and Language Processing, vol.15, no.6, 1932--1943, 2007.
• K. Yu and M. J. F. Gales (2006).
  Discriminative cluster adaptive training.
  IEEE Transactions on Audio, Speech and Language Processing, vol. 14, no.5, 1694--1703, 2006.
• K. Yu and L. Ji (2002).
  Karyotyping of CGH human metaphases using kernel nearest-neighbor algorithm.
  Cytometry, vol. 48, no.4, 202--208, 2002.
• K. Yu, L. Ji and X. Zhang (2002).
  Kernel nearest-neighbor algorithm.
  Neural Processing Letters, vol. 15, no. 2, 147--156, 2002.
• K. Yu, L. Ji, L. Wang and P. Xue (2001).
  How to optimize OCT image.
  Optics Express, vol. 9, no. 1, 24--35, 2001.

Peer Reviewed Conference Papers

• Milica Gasic, Filip Jurcicek, Blaise Thomson, Kai Yu and Steve Young. (2011)
  On-line policy optimisation of spoken dialogue system via live interaction with human subjects .
  IEEE ASRU 2011.
• A. W. Black, S. Burger, A. Conkie, H. Hastie, S. Keizer, O. Lemon, N. Merigaud, G. Parent, G. Schubiner, B. Thomson, J. D. Williams, K. Yu, S. Young and M. Eskenazi. (2011)
  Spoken Dialog Challenge 2010: Comparison of Live and Control Test Results .
  SIGDial 2011.
• F. Jurcicek, S. Keizer, M. Gasic, F. Mairesse, B. Thomson, K. Yu, and S. Young. (2011)
  Real user evaluation of spoken dialogue systems using Amazon Mechanical Turk.
  INTERSPEECH 2011.
• K. Yu and S. Young. (2011)
  Joint modelling of voicing label and continuous F0 for HMM based speech synthesis.
  ICASSP 2011.
• L. Jia, K. Yu and B. Xu. (2011)
  Structured precision modelling with Cholesky basis superposition for speech recognition.
  ICASSP 2011.
• B. Thomson, K. Yu, S. Keizer, M. Gasic, F. Jurcicek, F. Mairesse and S. Young. (2010)
  Bayesian dialogue system for the Let's Go spoken dialogue challenge.
  IEEE SLT 2010.
• B. Thomson, F. Jurcicek, M. Gasic, S. Keizer, F. Mairesse, K. Yu and S. Young. (2010)
  Parameter learning for POMDP spoken dialogue models.
  IEEE SLT 2010.
• K. Yu, B. Thomson and S. Young. (2010)
  From discontinuous to continuous F0 modelling In HMM-based speech synthesis.
  ISCA SSW7 2010.
• K. Yu, H. Zen, F. Mairesse and S. Young. (2010)
  Context adaptive training with factorized decision trees for HMM-based speech synthesis.
  INTERSPEECH 2010.
• M. Gales and K. Yu. (2010)
  Canonical state models for automatic speech recognition.
  INTERSPEECH 2010.
• F. Jurcicek, B. Thomson, S. Keizer, F. Mairesse, M. Gasic, K. Yu and S. Young. (2010)
  Natural Belief-Critic: a reinforcement algorithm for parameter estimation in statistical spoken dialogue systems.
  INTERSPEECH 2010.
• M. Gasic, F. Jurcicek, S. Keizer, F. Mairesse, B. Thomson, K. Yu and S. Young. (2010)
  Gaussian processes for fast policy optimisation of POMDP-based dialogue managers.
  SIGDial 2010.
• S. Keizer, M. Gasic, F. Jurcicek, F. Mairesse, B. Thomson, K. Yu and S. Young. (2010)
  Parameter estimation for agenda-based user simulation.
  SIGDial 2010.
• F. Mairesse, M. Gasic, F. Jurcicek, S. Keizer, J. Prombonas, B. Thomson, K. Yu and S. Young. (2010)
  Phrase-based statistical language generation using graphical models and active learning.
  ACL 2010.
• K. Yu, F. Mairesse and S. Young. (2010)
  Word-level emphasis modelling in HMM-based speech synthesis.
  ICASSP 2010.
• M. Gasic, F. Lefevre, F. Jurcicek, S. Keizer, F. Mairesse, B. Thomson, K. Yu, and S. Young. (2009)
  Back-off action selection in summary space-based POMDP dialogue systems.
  IEEE ASRU 2009.
• F. Lefevre, M. Gasic, F. Jurcicek, S. Keizer, F. Mairesse, B. Thomson, K. Yu, and S. Young. (2009)
  K-nearest neighbor Monte-Carlo control algorithm for POMDP-based dialogue systems.
  SIGDial 2009.
• F. Jurcicek, M. Gasic, S. Keizer, F. Mairesse, B. Thomson, K. Yu, and S. Young. (2009)
  Transformation-based learning for semantic parsing.
  INTERSPEECH 2009.
• K. Yu, T. Toda, M. Gasic, S. Keizer, F. Mairesse, B. Thomson and S. Young. (2009)
  Probabilistic modelling of F0 in unvoiced regions in HMM based speech synthesis.
  ICASSP 2009.
• F. Mairesse, M. Gasic, F. Jurcicek, S. Keizer, B. Thomson, K. Yu and S. Young. (2009)
  Spoken language understanding from unaligned data using discriminative classification models.
  ICASSP 2009.
• S. Keizer, M. Gasic, F. Mairesse, B. Thomson, K. Yu and S. Young. (2008)
  Modelling user behaviour in the HIS-POMDP dialogue manager.
  IEEE SLT 2008.
• C. K. Raut, K. Yu and M. J. F. Gales. (2008)
  Adaptive training using discriminative mapping transforms.
  INTERSPEECH, 2008.
• B. Thomson, K. Yu, M. Gasic, S. Keizer, F. Mairesse, J. Schatzmann and S. Young. (2008)
  Evaluating semantic-level confidence scores with multiple hypotheses.
  INTERSPEECH, 2008.
• B. Thomson, M. Gasic, S. Keizer, F. Mairesse, J. Schatzmann, K. Yu and S. Young. (2008)
  User study of the Bayesian Update of Dialogue State approach to dialogue management.
  INTERSPEECH, 2008.
• M. Gasic, S. Keizer, F. Mairesse, J. Schatzmann, B. Thomson, K. Yu and S. Young. (2008)
  Training and evaluation of the HIS POMDP dialogue system in noise.
  SIGDIAL 2008
• K. Yu, M. J. F. Gales and P. C. Woodland. (2008)
  Unsupervised discriminative adaptation using discriminative mapping transforms.
  ICASSP 2008
• X. Liu, W. Byrne, M. J. F. Gales, A. Gispert, M. Tomalin, P. C. Woodland and K. Yu. (2007)
  Discriminative language model adaptation for Mandarin broadcast speech transcription and translation.
  IEEE ASRU 2007
• M. J. F. Gales, F. Diehl, C. K. Raut, M. Tomalin, P. C. Woodland and K. Yu. (2007)
  Development of a phonetic system for large vocabulary Arabic speech recognition.
  IEEE ASRU 2007
• K. Yu, M. J. F. Gales and P. C. Woodland. (2007)
  Unsupervised training with directed manual transcription for recognizing Mandarin broadcast audio.
  INTERSPEECH 2007
• M. J. F. Gales, X. Liu, R. Sinha, P. C. Woodland, K. Yu, S. Matsoukas, T. Ng, K. Nguyen, L. Nguyen, J.-L. Gauvain, L. Lamel, A.Messaoudi. (2007)
  Speech recognition system combination for machine translation.
  ICASSP 2007
• M. Tomalin, M. J. F. Gales, X. Liu, K. C. Sim, R. Sinha, L. Wang, P. C. Woodland and K. Yu. (2007)
  Improving speech transcription for Mandarin-English translation.
  ICASSP 2007
• K. Yu and M.J.F. Gales (2006).
  Incremental adaptation using Bayesian inference.
  ICASSP 2006
• K. Yu and M.J.F. Gales (2005).
  Bayesian adaptation and adaptively trained systems.
  IEEE ASRU 2005
• G. Evermann, H. Y. Chan, M. J. F. Gales, B. Jia, D. Mrva, P. C. Woodland and K. Yu. (2005)
  Training LVCSR systems on thousands of hours of data.
  ICASSP 2005
• M. J. F. Gales, B. Jia, X. Liu, K. C. Sim, P. C. Woodland and K. Yu. (2005)
  Development of the CUHTK 2004 Mandarin conversational telephone speech transcription system.
  ICASSP 2005
• X. Liu, M. J. F. Gales, K. C. Sim and K. Yu. (2005)
  Investigation of acoustic modeling techniques for LVCSR systems.
  ICASSP 2005
• K. Yu and M.J.F. Gales (2004).
  Adaptive Training Using Structured Transforms.
  ICASSP 2004
• S. E. Tranter, K. Yu, G. Evermann and P. C. Woodland. (2004)
  Generating and evaluating segmentations for automatic speech recognition of conversational telephone speech.
  ICASSP 2004

Technical Reports

• K. Yu and M.J.F. Gales (2006).
  Bayesian adaptation and adaptive training.
  Technical Report CUED/F-INFENG/TR542
• K. Yu and M.J.F. Gales (2004).
  Discriminative cluster adaptive training.
  Technical Report CUED/F-INFENG/TR486
• S. E. Tranter, K. Yu, D. A. Reynolds, G. Evermann, D. Y. Kim and P. C. Woodland.
  An investigation into the interactions between speaker diarisation systems and automatic speech transcription.
  Technical Report CUED/F-INFENG/TR464

Contact Information