SPEAKER: XUEDONG HUANG
Senior Researcher, Speech Technology Group, Microsoft Research and
Affiliate Professor of Electrical Engineering, University of Washington
ABSTRACT:
SPEAKER BIO:
Xuedong served as the vice general chair of ICASSP '98, an associate editor
for IEEE Transaction on Speech and Audio processing from 1992 to 1996, and
a member of Pattern Analysis and Applications editorial board. He is a
Senior Member of IEEE, the co-author of the book Hidden Markov Models for
Speech Recognition, Edinburgh University Press, 1990, and more than 70
papers on spoken language technology. His awards include: National Education
Commission of China's 1987 Science and Technology Progress Award, IEEE
Signal Processing Society's 1993 Paper Award, and The Allen Newell Medal for
Research Excellence. Xuedong received his B.S. from Hunan Universityhttp://www.hunu.edu.cn/, M.S. from
Tsinghua University http://www.tsinghua.edu.cn/docse/eindex.html, and Ph.D. from the
University of Edinburgh http://www.ed.ac.uk/ .
Can We Handle Unrestricted Dialogs?
In his classic overview paper published by IEEE more than 20
years ago, Raj Reddy said we were far from being able to handle
unrestricted dialogs from a large population of speakers in
uncontrolled environments. What Raj discussed in that paper
still holds insights for us. In particular, we still face some
of the challenges Raj outlined: signal processing associated with
noise, telephone and speaker normalization, graceful user interface,
and labeled data bases. In this talk, I will review Microsoft's
efforts to build natural conversational systems, with particular
emphasis on speech recognition, speech synthesis, and dialog management.
Xuedong Huang joined Microsoft as a Senior Researcher to
head Microsoft's speech technology group in January 1993. He is
responsible for creating spoken language technologies used for
Microsoft's speech recognition and synthesis systems and Speech
API (SAPI). He is also an affiliate Professor of Electrical Engineering at
the University of Washington. Before joining Microsoft, Xuedong directed
the effort in developing the Sphinx-II speech recognition system at
Carnegie Mellon University.