Subject: RA Opportunities in Speech Recognition
Date: Tue, 21 Jan 1997 18:56:08 +0000 (GMT)
From: Steve Young <sjy@eng.cam.ac.uk>
To: elsnet-list@let.ruu.nl

          Cambridge University Engineering Department

                   Research Associates in

             Large Vocabulary Speech Recognition


Applications are invited to join a team developing the HTK Large
Vocabulary Speech Recognition System.  The persons appointed will
contribute to the general development of the system and conduct
research into methods for improving performance on transcribing
spontaneous speech including telephone and broadcast news.  The work
will involve participation in the international benchmark evaluations
conducted by the US National Institute of Standards and Technology
(NIST) and attendance at the associated workshops.

The HTK-based large vocabulary continuous speech recognition system
has a vocabulary of 65,000 words.  It uses cross-word context
dependent HMM models, N-gram language models and incremental
adaptation.  Work in progress includes acoustic modelling, automatic
identification of speaker and environment, rapid adaptation, and
language modelling.

Candidates should have a good first degree in a relevant discipline
and preferably have a higher degree. Preference will be given to
candidates with experience in speech recognition, C programming, and
UNIX.  Familiarity with the HTK Toolkit is also a distinct advantage.

Prospective candidates should send a CV to 

   Professor Steve Young              Email:  sjy@eng.cam.ac.uk
   Speech, Vision and Robotics Group  Office: +44-1223-332654
   Engineering Department             Fax:    +44-1223-332662
   Trumpington Street                 Mobile: 0850-305131
   Cambridge, CB2 1PZ, England        WWW: http://svr-www.eng.cam.ac.uk