free stats

Yun-Hsuan Sung (宋雲軒)

Research Scientist
Google Research
Email: yhsung(at)stanford.edu

I have graduated from Stanford University and join Google as a research scientist.

I was a PhD student in the Department of Electrical Engineering at Stanford University. I work with Prof. Dan Jurafsky and Chris Manning on automatic speech recognition in Natural Language Processing Group and Speech Lab. I am insterested in applying machine learning and statistical graphical models to speech and language processing. I came from Taiwan and got my BS degree from the Department of Electrical Engineering at National Taiwan University in 2002.

Research Interests:

Speech and language processing, machine learning, graphical models, information retrieval

Experience:

2010.03 ~ present Research Scientist
Google Inc, CA, USA
2005.09 ~ 2010.03 Research Assistant
NLP Group and Speech Lab, Stanford University, CA, USA
2008.06 ~ 2008.04 Summer Intern
Speech Research Group, Google Research, CA, USA
2007.01 ~ 2007.04 Teaching Assistant
CS224S/LING281 Speech Recognition and Synthesis, Stanford University, CA, USA
2006.06 ~ 2006.09 Summer Intern
Speech Research Group, Microsoft Research, WA, USA

Publications:

Recognition of Multilingual Speech in Mobile Applications
Hui Lin, Jui-Ting Huang, Francoise Beaufays, Brian Strope, and Yun-Hsuan Sung
To appear in ICASSP 2012, Kyoto, Japan
Deploying Google Search by Voice in Cantonese
Yun-Hsuan Sung, Martin Jansche, and Pedro Moreno
In proceedings of Interspeech 2011, Florence, Italy
Recognizing English Queries in Mandarin Voice Search
Hung-An Chang, Yun-Hsuan Sung, Brian Strope, and Francoise Beaufays
In proceedings of IEEE ICASSP 2011, Prague, Czech Republic
Hidden Conditional Random Fields for Speech Recognition
PhD dissertation, Stanford University, 2010.
Hidden Conditional Random Fields for Phone Recognition [pdf]
Yun-Hsuan Sung and Dan Jurafsky
In proceedings of IEEE ASRU 2009, Merano, Italy
Revisiting Graphemes with Increasing Amounts of Data [pdf]
Yun-Hsuan Sung, Thad Hughes, Francoise Beaufays, and Brian Strope
In proceedings of IEEE ICASSP 2009, Taipei, Taiwan
Maximum Conditional Likelihood Linear Regression and Maximum a Posteriori for Hidden Conditional Random Fields Speaker Adaptation [pdf]
Yun-Hsuan Sung, Constantinos Boulis, and Dan Jurafsky
In proceedings of IEEE ICASSP 2008, Las Vegas, USA
Regularization, Adaptation, and Non-Independent Features Improve Hidden Conditional Random Fields for Phone Classification [pdf]
Yun-Hsuan Sung, Constantinos Boulis, Christopher Manning, and Dan Jurafsky
In proceedings of IEEE ASRU 2007, Kyoto, Japan
Detection of Word Fragments in Mandarin Telephone Conversation [pdf]
Cheng-Tao Chu, Yun-Hsuan Sung, Zhao Yuan, and Dan Jurafsky
In proceedings of Interspeech 2006, Pittsburgh PA, USA