[Elsnet-list] Open Position: Research Associate in Automatic Speech Recognition

Thomas Hain T.Hain at dcs.shef.ac.uk
Sun Dec 3 21:25:44 CET 2006


                          UNIVERSITY OF SHEFFIELD
                       DEPARTMENT OF COMPUTER SCIENCE
                     SPEECH AND HEARING RESEARCH GROUP

            RESEARCH ASSOCIATE IN AUTOMATIC SPEECH RECOGNITION (ASR)
                              REF:  PR2672

BRIEF OUTLINE

The Department of Computer Science has a position for a research
associate to work on automatic speech recognition (ASR). The
position is funded by the EC as part of the AMIDA project (see
http://www.amiproject.org). The project is a successor to the AMI
project and aims to research augmentative technology that allows
increasing efficiency of business meetings.
In AMIDA, special focus is given to technology relevant for
people that are not physically collocated. More than ten
international institutions work jointly on a multitude of tasks,
ranging from video analysis to behavioural modelling and audio
event detection to question answering (for more details please
visit the website). Naturally, automatic speech recognition forms
a core part of the analysis of meetings. Its output is needed for
presentation and further higher level processing.

The development of the AMIDA ASR system is shared between six
partners under the leadership of the University of Sheffield. The
work will use, and extend, the existing state-of-the-art large
vocabulary recognition system developed under AMI, which was
shown in the recent U.S. NIST RT06s ASR evaluations to be very
competitive.

This is an opportunity to work in a well connected international
team with world-leading reputations in speech recognition
research for meetings, and to be embedded in the Speech and
Hearing (SpandH) research group
(http://www.dcs.shef.ac.uk/spandh) at the University of
Sheffield. SpandH is well known internationally for its research,
and is one of the largest research groups in speech and hearing
in the UK.

There are excellent opportunities for publications, travel and conference
visits. The group also has excellent research facilities. For suitably
qualified candidates possibilities exist to contribute to teaching, for
example as part of the MSc in Human Language Technologies. Candidates must
have a good first degree (or equivalent experience) in a relevant
discipline and preferably have a higher degree. Knowledge of Unix type
operating systems and programming in C/C++ is required. In addition,
experience in one or more of the following technical areas would be a
distinct advantage:

* Acoustic modelling for automatic speech recognition
* Statistical pattern processing
* Microphone array signal processing
* The Hidden Markov Model Toolkit (HTK)

Candidates are expected to work well in research teams and contribute to
research and advancement of the SpandH research group.

FURTHER INFORMATION

INFORMAL ENQUIRIES:  Informal enquiries may be directed to Dr Thomas Hain
(Tel. +44 (0)114 222 1836 or email: t.hain at dcs.shef.ac.uk).
FURTHER INFORMATION: This post is available from 1 January 2007 for a
period of 30 months.
SALARY:  24,161 - 31,5 UKP per annum.
CLOSING DATE:  2 January 2007.

            A MORE DETAILED DESCRIPTION CAN BE FOUND ON
                       www.sheffield.ac.uk/jobs
                   WITH JOB REFERENCE NUMBER PR2672.


Best regards 
Thomas Hain 


More information about the Elsnet-list mailing list