[opensource] [AI] CSE Distinguished Lecture on Speech Recognition

Michael S. Yanovich yanovich.1 at osu.edu
Fri Nov 5 16:58:07 EDT 2010


I think this looks interesting. I think others here might appreciate
this lecture.

-------- Original Message --------
Subject: [AI] CSE Distinguished Lecture on Speech Recognition
Date: Fri, 05 Nov 2010 11:49:24 -0400
From: DeLiang Wang <dwang at cse.ohio-state.edu>
To: CSE Faculty <faculty at cse.ohio-state.edu>, CSE Grads
<grads at cse.ohio-state.edu>, CSE Undergrads
<undergrads at cse.ohio-state.edu>, ai at cse.ohio-state.edu, Announcements
and programs of the Center for Cognitive Science at OSU
<cog-announce at lists.service.ohio-state.edu>

CSE Distinguished Seminar Series presents

Douglas O'Shaughnessy
INRS-EMT, University of Quebec

"Modern Methods of Automatic Speech Recognition (ASR): How and Why"

Time: 3:30PM, Nov 16, 2010
Place: 480 Dreese Labs

Refreshments will be served prior to the seminar

Abstract:
Researchers have been developing ways to convert speech into text since
the advent of serious digital computers, i.e., for about 50 years now.
To help set a course for future improvements in ASR, it helps to
understand the progression of major developments in the field of ASR,
not only what the algorithms are but why they developed the way they
did. We will examine the fundamentals of basic methodology of modern
ASR, i.e., A/D conversion, spectral analysis, simple auditory models,
efficient representations such as linear predictive coding (also used
for coding in cellular phones) and MFCC, hidden Markov models, training
issues, and databases, from the points of view of efficiency and
accuracy. The approach is one of pattern recognition of a well-known but
highly-variable source, i.e., the human vocal tract. ASR must take
advantage of knowing what the source is, but also exploit the
communicative nature of speech signals. Rendering the problem of ASR
more complex is its interdisciplinary nature, involving acoustics,
engineering, computer science and linguistics. The talk will assume
basic knowledge of engineering and computers, but not assume a
specialized knowledge of speech processing.

Speaker Biography:/

Douglas O'Shaughnessy
<http://externe.emt.inrs.ca/users/spchwww/English/persons/dougo.html> is
professor and program director at INRS-EMT (formerly,
INRS-Telecommunications), a constituent of the University of Quebec, in
Montreal, Canada. He is also an adjunct professor at McGill University
in the Department of Electrical and Computer Engineering. He was
educated at MIT (B.Sc. and M.Sc. in 1972; Ph.D. in 1976). His interests
and research include automatic speech synthesis, analysis, coding,
enhancement, and recognition. He has served as an Associate Editor for
IEEE Transactions on Speech and Audio Processing, and the Journal of the
Acoustical Society of America. He was the General Chair of the 2004
International Conference on Acoustics, Speech and Signal Processing
(ICASSP) in Montreal, Canada. He is the author of the textbook Speech
Communications: Human and Machine (2nd edition in 2000 by IEEE press).
He is a Fellow of the Acoustical Society of America and of IEEE.

/

Host: DeLiang Wang <mailto:dwang at cse.ohio-state.edu>



_______________________________________________
AI mailing list
AI at mail.cse.ohio-state.edu
http://mail.cse.ohio-state.edu/mailman/listinfo/ai

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 899 bytes
Desc: OpenPGP digital signature
Url : http://mail.cse.ohio-state.edu/pipermail/opensource/attachments/20101105/718c2ab1/signature.bin


More information about the Opensource mailing list