I obtained my bachelor and master degrees in Department of Electrical and Electronic Engineering at Hong Kong University of Science Technology (HKUST). My master thesis is supervised by Prof. Manhung Siu. The thesis is about improving speech recognition in impulse noise. That requires re-formulation of the conventional Viterbi algorithm.
After I left HKUST, I became a Speech Scientist in Speechworks (now Scansoft) at where I improved performance of speech recognition for Cantonese, Singaporean English, Australian English and American English in Speechworks 6.5 recognizer, Open Speech Recognizer (OSR) 1.1 and 2.0. I also involved in speech projects serving clients such as Singtel and Qantas.
From 2004 to 2006, I was working with Prof. Alex Rudnicky in the Computer Science Department of Carnegie Mellon University (CMU) as a Senior Research Programmer. I put most of my effort to improve open-source Sphinx, a speaker-independent large vocabulary continuous speech recognizer developed by speech group of CMU. I was working on Sphinx 3.X, a collection of LVCSR decoders (include so called s3slow (sphinx 3.0) and s3fast) and SphinxTrain, a collection of tools to build acoustic models for state of the art speech recognition systems. I also worked in DARPA funded projects CALO (speech processing) and GALE (speech recognition effort).
I am now at Scanscout (stealth mode) and largely involved in its speech recognition, natural language processing, information retrieval and video processing effort.
Feed the above message into Festival in the TTS mode, you will get it. :-)
"Yu Chung" is my Chinese first name. My first few papers used it, but most of the people work with me usually called me Arthur so sometimes the author names of my papers messed up.
My master thesis (Don't remind me it is poor, I know. :-))
B. Langner, R. Kumar, A. Chan , L. Gu, A. W. Black, "Generating Time-Constrained Audio Presentations of Structured Information", To be appeared in Interspeech 2006, Pittsburgh, USA
D. Huggins-Daines, M. Kumar, A. Chan , A. W. Black, R. Mosur, A. I. Rudnicky , "PocketSphinx: A Free, Real-time Continuous Speech Recognition System for Hand-held Devices", ICASSP2006, France ( ps )
A. Chan , R. Mosur and A. I. Rudnicky, "On Improvements of CI-based GMM Selection", in Interspeech 2005, Portugal. ( pdf ).
R. Zhang, Z. Al Bawab, A. Chan , A. Chotiomongkol, D. Huggins-Daines, A. I. Rudnicky, "Investigations on Ensemble Based Semi-Supervised Acoustic Model Training", in Interspeech2005, Portugal. ( pdf )
A. Chan , and M. Siu, "Efficient Computation of the Frame-based Extend Union Model and its Application against Partial Temporal Corruptions'', in Computer Speech and Language V19 p.301-319. ( ps )
A. Chan , J. Sherwan, R. Mosur and A. I. Rudnicky, "Four-Level Categorization Scheme of Fast GMM Computation Techniques in Large Vocabulary Continuous Speech Recognition Systems", International Conference of Speech and Language Processing 2004, Korean. ( ps ).
S. Banerjee, J. Cohen, T. Quisel, A. Chan , Y. Patodia, Z. Al Bawab, R. Zhang, A. Black, R. Stern, R. Rosenfeld, A. I. Rudnicky, "Creating Multi-Modal, User-Centric Records of Meetings with the Carnegie Mellon Meeting Recorder Architecture", NIST Meeting Recognition Workshop of ICASSP 2004.
M. Siu and A. Chan , "A Robust Viterbi Algorithm Against Impulsive Noise with Application for Speech Recognition'', accepted by IEEE Transaction on Speech and Audio Proceeding. ( (Final Manuscript Version) ps )
Brian Mak, M Siu, Mimi Ng, Y-C Tam, Y. C. Chan , K-W Chan, K-Y Leung, S. Ho, J. Wong and J. Lo, "PLASER: Pronunciation Learning via Automatic Speech Recognition'', Proc. HLT-NAACL 2003 Workshop on Building Educational Applications using Natural Language Processing, Edmonton, Canada, May 31, 2003, pages 23-29. ( pdf )
M. Siu, Y. C. Chan , "Robust Speech Recognition Against Short-Time noise'', International Conference on Spoken Language Processing, Volume 2, pages 1049-1052, Sep 2002. ( ps )
M. Siu, Y. C. Chan , "A Modified Viterbi Algorithm that Skips K-frames'', European Conference on Speech Comm. and Tech. 2001 ( ps )
Y. C. Chan , M. Siu and K.W Mak, "Pruning of the State-Tying Tree using the Bayesian Information Criterion with Multiple Mixtures'', Proc. International Conference on Spoken Language Processing, volume 4, pages 294-297, Oct 2000 ( ps )
My master thesis presentation gives me very painful memory. I include my presentation here to remind myself not to write lengthy presentations again. I will regard my presentation in CMU is the most successfull one for this series of talk.
Thesis presentation Presentation in Speechworks Presentation in CMUI put most of those stuffs to another page
.