[Elsnet-list] PhD Position in Speech Recognition for under-resourced languages (K.U.Leuven, Belgium)

Dirk Van Compernolle (ESAT) compi at telenet.be
Sat Apr 16 11:38:45 CEST 2011


*Speech Recognition for Under-resourced Languages

K.U.Leuven - ESAT, Belgium
*
------------------------------------------------------------------------

*Project Description*

Today's speech recognition systems require hundreds of hours of example 
data for training the acoustic models.  While such large corpora are 
available for the major languages, this is not the case for smaller 
languages, making them "under-resourced". One of the underlying reasons 
for this data hungriness is that the dimensionality of feature vectors 
used in state-of-the-art speech recognition systems (typically in the 
range 30-40) is much larger than the intrinsic dimensionality of speech 
which is estimated to be 7-10 only. Efforts to make the intrinsic 
dimensionality smaller have been largely futile as the constraints are 
too complex for our by and large linear techniques. This inefficiency in 
basic representation, combined with other inefficiencies in mainstream 
context-dependent modeling makes that the hundreds of thousands 
parameters that constitute an acoustic model are largely redundant.

The objective of this project is to apply novel mathematical techniques 
(e.g. spectral clustering) that can capture constraints - not in the 
feature space - but in the model space, i.e. in the underlying HMM 
parameters. Such constraints will lead to lesser requirements on the 
size of the training databases and should increase robustness in all 
situations where we don't have large corpora available, such as speaker 
adaptation, accent adaptation or modeling of under-resourced languages. 
Apart from general principles, two test cases will be  be studied in 
more detail : i) "Afrikaans", for which data from Dutch and Flemish can 
be reused; ii) languages form the Bantu family as spoken in South Africa 
for which we can only bootstrap from a wide set of rather unrelated 
languages.


*Qualifications*

Candidates ideally have a university degree in engineering, computer 
science or applied mathematics. Skills and experience in any of the 
following areas are welcomed:

    * strong mathematical background (linear algebra and/or statistical
      parameter estimation)
    * speech recognition and speech modeling
    * some familiarity with Dutch or Afrikaans
    * computational skills (MATLAB, C, UNIX, Python)

*Position*

Within this project there is funding for a 4yr Ph.D. scholarship.  
Alternatively we will also accept applications for a 2 yr junior 
post-doc with significant relevant experience.

*Project Partners*

This research will be carried out the K.U.Leuven, Belgium in the context 
of the AMODA project in collaboration with Council for Scientific and 
Industrial Research (CSIR), Pretoria, South Africa.

*Contact*

Interested candidates should contact Prof. Dirk Van Compernolle  "compi 
AT esat.kuleuven.be"


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.elsnet.org/pipermail/elsnet-list/attachments/20110416/b30660b0/attachment.htm 


More information about the Elsnet-list mailing list