Share this page
  • Share this page on Twitter Share this page on Facebook Share this page on Digg Share this page on Del.icio.us Read the Inside Microsoft Research blog
  • E-mail this page Print this page
  • RSS feeds
Home  > People > Xiaodong He
Xiaodong He
   

Xiaodong He is a Researcher in the Speech Research Group.

My homepage is moved to another place, please visit my new site for up-to-date information.

Background and Interests

Xiaodong He received the BS degree from Tsinghua University (Beijing) in 1996, the MS degree from Chinese Academy of Sciences in 1999, and the PhD degree from University of Missouri - Columbia in 2003. His current research interests include statistical pattern recognition and machine learning, machine translation, speech recognition and translation, natural language processing, and information retrieval. He is a senior member of IEEE and a member of ACL. He is a member of Sigma Xi. He was a Co-Chair of the NIPS 2008 Workshop on Speech and Language, and the lead guest editor of the IEEE J-STSP special issue on Statistical Learning Methods for Speech and Language Processing. He is an associate editor of IEEE Signal Processing Magazine.

In 2006, he joined Microsoft Research (MSR), Redmond, WA, where he is currently a researcher in the Speech Research Group. at MSR, he has worked in the speech and natural language processing areas, including speech recognition and translation, machine translation, information retrieval and web search, and model optimization techniques. In the 2008 NIST Open MT Evaluation, he and colleagues submitted the MSR-NRC-SRI entry, which gave the best Chinese-to-English result in the Common-Data Track.

Prior to joining MSR, he was with Microsoft Speech Components Group from 2003 to 2006, where he engaged in a wide range of speech recognition research and development activities including large vocabulary discriminative training, confidence measure, and speaker adaptation.

He has authored/coauthored more than 40 papers and one book in these areas. His work has been incorporated into a variety of Microsoft speech and translation products.

Special Issue in IEEE Journal of Selected Topics in Signal Processing

We edited a Special Issue on Statistical Learning Methods for Speech and Language Processing, which was published in the IEEE Journal of Selected Topics in Signal Processing in December, 2010.

NIPS 2008 Workshop

We orgnized the NIPS 2008 workshop Speech and Language: Learning-based Methods and Systems. It covers a variety of advanced topics in the Speech and Language Processing areas. More details can be found at NIPS08 WSL(a)

Publications

Book

  • Xiaodong He and Li Deng, 2008. Discriminative Learning for Speech Recognition: Theory and Practice, Morgan & Claypool Publishers, 2008. ISBN: 1598293087 (order from Amazon.com)

Journals

  • Xiaodong He and Li Deng. 2011. Speech Recognition, Machine Translation, and Speech Translation – A Unified Discriminative Learning Paradigm. IEEE Signal Processing Magazine, September 2011 (draft)
  • Xiaodong He, Mei Yang, Jianfeng Gao, Patrick Nguyen, Robert Moore. 2009. Improved Monolingual Hypothesis Alignment for Machine Translation System Combination. ACM Transactions on Asian Language Information Processing -- special issue on Machine Translation of Asian Languages, June 2009 (draft)
  • Xiaodong He, Li Deng, Wu Chou, 2008. Discriminative Learning in Sequential Pattern Recognition –– A Unifying Review for Optimization-Oriented Speech Recognition. IEEE Signal Processing Magazine, September 2008 (draft)
  • Dong Yu, Li Deng, Xiaodong He, and Alex Acero, 2008. Large-Margin Minimum Classification Error Training: A Theoretical Risk Minimization Perspective. Computer Speech and Language, Vol. 22, October 2008 (draft)
  • Xiaodong He and Li Deng, 2007. A New Look at Discriminative Training for Hidden Markov Models. Invited paper, Pattern Recognition Letters, Vol. 28, August 2007 (draft)
  • Xiaodong He and Yunxin Zhao, 2007. Prior Knowledge Guided Maximum Expected Likelihood based Model Selection and Adaptation for Nonnative Speech Recognition. Computer Speech and Language, Vol. 21, April 2007 (draft)
  • Xiaodong He and Yunxin Zhao, 2003 Fast Model Selection Based Speaker Adaptation for Nonnative Speech. IEEE Transaction on Speech and Audio Processing, Vol. 11, July 2003 (draft)

Conferences

  • Xiaodong He and Li Deng, 2011. Robust Speech Translation by Domain Adaptation. Interspeech August 2011 (to appear) (draft)
  • Amittai Axelrod, Xiaodong He, and Jianfeng Gao, 2011. Domain Adaptation via Pseudo In-Domain Data Selection. EMNLP July 2011 (to appear) (draft)
  • Xiaodong He, Li Deng, and Alex Acero, 2011. Why Word Error Rate Is Not A Good Metric For Speech Recognizer Training For The Speech Translation Task? ICASSP May 2011 (draft)
  • Yaodong Zhang, Li Deng, Xiaodong He, Alex Acero, 2011. A Novel Decision Function and the Associated Decision-Feedback Learning For Speech Translation. ICASSP May 2011 (draft)
  • Jianfeng Gao, Xiaodong He, and Jian-Yun Nie, 2010. Clickthrough-Based Translation Models for Web Search: from Word Models to Phrase Models. CIKM October 2010 (draft)
  • Xiaodong He and Kristina Toutanova, 2009. Joint Optimization for Machine Translation System Combination. EMNLP August 2009 (draft)
  • Chi-Ho Li, Xiaodong He, Yupeng Liu and Ning Xi, 2009. Incremental HMM Alignment for MT System Combination. ACL. August 2009 (draft)
  • Yong Zhao and Xiaodong He, 2009. Using N-gram based Features for Machine Translation System Combination. NAACL-HLT May 2009 (draft)
  • Xiaodong He, Mei Yang, Jianfeng Gao, Patrick Nguyen, Robert Moore, 2008. Indirect-HMM-based Hypothesis Alignment for Combining Outputs from Machine Translation Systems. EMNLP. October 2008 (draft)
  • X. He, J. Gao, C. Quirk, P. Nguyen, A. Menezes, R. Moore, K. Toutanova, M. Yang, W. Dolan, M. Li, C.-H. Li, D. Zhang, L. Jiang, C. Niu, M. Zhou, G. Foster, R. Kuhn, J. Zheng, W. Wang, N. F. Ayan, D. Vergyri, N. Scheffer, A. Stolcke, 2008. The MSR-NRC-SRI MT System for NIST Open Machine Translation 2008 Evaluation. the 2008 NIST Open Machine Translation Evaluation Workshop. March 2008 (draft)
  • X. He, J. Gao, C. Quirk, P. Nguyen, A. Menezes, R. Moore, K. Toutanova, M. Yang, W. Dolan, M. Li, C.-H. Li, D. Zhang, L. Jiang, C. Niu, M. Zhou, 2008. The MSR-MSRA MT System for NIST Open Machine Translation 2008 Evaluation. the 2008 NIST Open Machine Translation Evaluation Workshop. March 2008 (draft)
  • Xiaodong He, 2007. Using Word-Dependent Transition Models in HMM based Word Alignment for Statistical Machine Translation. ACL07 2nd SMT workshop. (draft)
  • Patrick Nguyen, Milind Mahajan and Xiaodong He, 2007. Training Non-Parametric Features for Statistical Machine Translation. ACL 07 2nd SMT workshop. (draft)
  • Masaki Itagaki, Takako Aikawa, Xiaodong He, 2007. Automatic Validation of Terminology Translation Consistency with Statistical Method. MT Summit XI. (draft)
  • Dong Yu, Li Deng, Xiaodong He, Alex Acero, 2007. Large-Margin Minimum Classification Error Training For Large-Scale Speech Recognition Tasks. ICASSP. (draft)
  • Qiang Fu, Xiaodong He, Li Deng, 2007. Phone-Discriminating Minimum Classification Error (P-MCE) Training for Phonetic Recognition. Interspeech. (draft)
  • X. He, A. Menezes, C. Quirk, A. Aue, S. Corston-Oliver, JF. Gao, and P. Nguyen, 2006. Microsoft Research Treelet Translation System: NIST MT Evaluation 06. the 2006 NIST Open Machine Translation Evaluation Workshop. (draft)
  • Xiaodong He, Li Deng, and Wu Chou, 2006. A Novel Learning Method for Hidden Markov Models in Speech and Audio Processing. IEEE MMSP. (draft)
  • Xin Lei, Jon Hamaker, and Xiaodong He, 2006. Robust Feature Space Adaptation For Telephony Speech Recognition. InterSpeech (draft)
  • Yu Dong, Li Deng, Xiaodong He, and Alex Acero, 2006. Use of Incrementally Regulated Discriminative Margins in MCE Training for Speech Recognition. InterSpeech (draft)
  • Xiaodong He and Yunxin Zhao, 2004. Prior knowledge guided MEL based model selection and adaptation for nonnative speech recognition. ICASSP (draft)
  • Xiaodong He and Wu Chou, 2003. Minimum Classification Error Linear Regression for Acoustic Model Adaptation of Continuous Density HMMs. ICASSP (draft)
  • Xiaodong He and Wu Chou, 2003. Minimum Classification Error (MCE) Model Adaptation of Continuous Density HMMs. In Proceedings of European Conf. on Speech Communication and Technology. (draft)
  • Wu Chou and Xiaodong He, 2003. Maximum a Posteriori Linear Regression (MAPLR) Variance Adaptation for Continuous Density HMMs. In Proceedings of European Conf. on Speech Communication and Technology. (draft)
  • Xiaodong He and Yunxin Zhao, 2002. Maximum Expected Likelihood Based Model Selection and Adaptation for Nonnative English Speakers. In Proceedings of Int'l Conf. on Spoken Language Processing. (draft)
  • Xiaodong He and Yunxin Zhao, 2002. Fast model adaptation and complexity selection for nonnative English speakers. ICASSP (draft)
  • Xiaodong He and Yunxin Zhao, 2001. Model Complexity Optimization for Nonnative English Speakers. In Proceedings of European Conf. on Speech Communication and Technology. (draft)
  • Yunxin Zhao, Xiao Zhang, Xiaodong He, Laura Schopp, 2000. A Combined Adaptive and Decision Tree Based Speech Separation Technique for Telemedicine Applications. In Proceedings of Int'l Conf. on Spoken Language Processing. (draft)
  • Xiaodong He, Jian Liu, Jianlai Zhou, Tiecheng Yu, 1999. Research on Speech Units Modeling in Continuous Speech Recognition. In Proceedings of European Conf. on Speech Communication and Technology. (draft)
  • Jianlai Zhou, Xiaodong He, Tiecheng Yu, Fuyuan Mo, 1999. A New Hybrid Structure of Speech Recognizer Based on HMM and Neural Network. In Proceedings of European Conf. on Speech Communication and Technology. (draft)
  • Jian Liu, Xiaodong He, Fuyuan Mo, Tiecheng Yu, 1999. Study on Tone Classification of Chinese Continuous Speech in Speech Recognition System. In Proceedings of European Conf. on Speech Communication and Technology. (draft)
  • Xiaodong He, Jian Liu, Tiecheng Yu, 1999. Research on Segmentation and Labeling of Speech Corpora. In Proceedings of Oriental COCOSDA, International Workshop on East-Asian Language Resource and Evaluation. (draft)
  • Xiaodong He, Jianlai Zhou, Jian Liu, Tiecheng Yu, 1999 A Speaker independent vocabulary extensible Chinese speech recognition system In Proceedings of International Symposium on Machine Translation and Computer Language Information Processing (draft)

Thesis

  • Xiaodong He, 2003. Model Selection based Speaker Adaptation and its Application to Nonnative Speech Recognition. PhD dissertation, University of Missouri-Columbia, 2003

Others

  • Xiaodong He and Li Deng, 2007. Discriminative Learning in Speech Recognition. Technical Report of Microsoft Research (MSR-TR-2007-129). Oct 2007.
  • Xiaodong He, 2004. Recognition Confidence and Threshold Tuning. Feature article, Microsoft Speech Server Newsletter July 2004

 

Patents

  • Use of incrementally regulated discriminative margins in mce training for speech recognition. with Dong Yu, Li Deng and Alex Acero.
  • Speech Models Generated Using Competitive Training, Asymmetric Training, And Data Boosting. with Jian Wu.
  • Segment Discriminating Minimum Classification Error Pattern Recognition. with Yu Dong, Li Deng, and Alex Acero.
  • Confidence Threshold Tuning, with Li Jiang, Julian Odell. and Wei Zhang. (pending)
  • Speech Recognition Using Adaptation and Prior Knowledge. with Xin Lei, Jonathan Hamaker, and Patrick Nguyen. (pending)
  • Identifying Language Origin of Words, with Min Chu, Yining Chen, Shiun-Zu Kuo, Kevin Feige. Yifan Gong, and Megan Riley. (pending)
  • A new method of discriminative training for hidden Markov models. with Li Deng. (pending)
  • Using Word Dependent Transition Models in HMM based Word Alignment for Statistical Machine Translation. (pending)
  • Automatic Validation of Terminology Translation Consistency and Its Applications. with Masaki Itagaki, Takako Aikawa. (pending)
  • A Generic Framework For Large Margin MCE in Speech Recognition. with Li Deng. (pending)
  • HMM Alignment for Combining Translation Systems. with Mei Yang, Jianfeng Gao, Patrick Nguyen, Robert Moore. (pending)
  • Using Machine Translation Technologies to help Foreign Language Education. with Alex Acero and Sebastian de la Chica. (pending)

 

Academic Services

  • Associate Editor, IEEE Signal Processing Magazine
  • Guest Editor, Special Issue in IEEE Journal of Selected Topics in Signal Processing, on Statistical Learning Methods for Speech and Language Processing
  • Co-Chair, NIPS 2008 Workshop on Speech and Language: Learning-Based Methods and Systems, Whistler, BC, Canada, 2008
  • Program Committee Member: IEEE International Conference on Semantic Computing (ICSC) 2007
  • Technical Committee Member: International Symposium on Chinese Spoken Language Processing (ISCSLP) 2008
  • Program Committee Member: The Int’l Conf. on Internet & Multimedia Systems & Applications, 2005, 2006, 2007
  • Reviewer: IEEE Transactions on Speech and Audio Processing, IEEE Signal Processing Magazine, IEEE Signal Processing Letters, IEEE Transactions on Computer, Speech Communication, Pattern Recognition, Pattern Recognition Letters, ICASSP, Interspeech, ASRU, HLT, NAACL, ACL, EMNLP, MT Summit

 

Honors and Awards

  • Gold Star Award from Microsoft, 2005
  • Patent awards from Microsoft in 2005-2008
  • Member of Sigma Xi since 2002
  • Award of student author grant, Int'l Conf. on Spoken Language Processing, 2002
  • Outstanding Academic Achievement Award, University of Missouri, 2001
  • Second Prize, the 14th "Challenge Cup" Sci & Tech Innovation Competition, Tsinghua University, 1996
  • Second Prize, the 13th "Challenge Cup" Sci & Tech Innovation Competition, Tsinghua University, 1995
  • Tsinghua – AIWA outstanding Sci. & Tech. Activities Award, 1995

E-mail: xiaohe@microsoft.com
U.S.Mail: Microsoft Corporation, One Microsoft Way, Redmond WA, 98052-6399, USA
Tel: (425) 706-4939
Fax: (425) 706-7329 (This is the main MS FAX number so make sure to send documents to Xiaodong He's attention)

Speech Research Group's home page.

Natural Language Processing group's home page.