Show simple item record

dc.contributor.authorBell, Peter
dc.contributor.authorKing, Simon
dc.date.accessioned2007-09-18T09:59:42Z
dc.date.available2007-09-18T09:59:42Z
dc.date.issued2007
dc.identifier.citationPeter Bell and Simon King. Sparse gaussian graphical models for speech recognition. In Proc. Interspeech 2007, Antwerp, Belgium, August 2007.en
dc.identifier.urihttp://hdl.handle.net/1842/1995
dc.description.abstractWe address the problem of learning the structure of Gaussian graphical models for use in automatic speech recognition, a means of controlling the form of the inverse covariance matrices of such systems. With particular focus on data sparsity issues, we implement a method for imposing graphical model structure on a Gaussian mixture system, using a convex optimisation technique to maximise a penalised likelihood expression. The results of initial experiments on a phone recognition task show a performance improvement over an equivalent full-covariance system.en
dc.format.extent128941 bytes
dc.format.mimetypeapplication/pdf
dc.language.isoenen
dc.subjectspeech technologyen
dc.titleSparse gaussian graphical models for speech recognition.en
dc.typeConference Paperen


Files in this item

This item appears in the following Collection(s)

Show simple item record