[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Speech separation challenge problem
SPEECH SEPARATION CHALLENGE!
Do you have an algorithm for speech separation, or do you just fancy a
new challenge? If so, you should take part in the 1st large-scale global
comparison of techniques for separating and recognising speech. Results
will be presented at a special session of Interspeech 2006
taking place in Pittsburgh (USA) from 17-21 September, 2006.
The task is to recognise speech from a target talker in the presence of
stationary noise or other speech. You will be provided with lots of
training data. Only one signal per mixture is provided (i.e. the task is
"single microphone"). Sentences have a simple structure to ensure the
overhead of building an automatic speech recognition system is kept small.
We welcome contributions from the widest possible range of approaches to
the speech separation/recognition problem, including well-known existing
algorithms as well as novel techniques.
For full details of the challenge and to download training/test sets, see
* April 7th 2006: submit your 4-page Interspeech paper for reviewing in
the normal way
* June 9th 2006: notification of acceptance
Martin Cooke (University of Sheffield, UK)
Te-Won Lee (UCSD, USA)