Publication Details
The subspace Gaussian mixture model-A structured model for speech recognition
Burget Lukáš, doc. Ing., Ph.D. (DCGM FIT BUT)
Agarwal Mohit (IIIT)
Akyazi Pinar (UBOGAZ)
Ghoshal Arnab (UEDIN)
Glembek Ondřej, Ing., Ph.D. (DCGM FIT BUT)
Goel Nagendra K. (GOVIVACE)
Karafiát Martin, Ing., Ph.D. (DCGM FIT BUT)
Rastrow Ariya (JHU)
Rose Richard (MCGILL)
Schwarz Petr, Ing., Ph.D. (DCGM FIT BUT)
and others
- http://www.fit.vutbr.cz/research/groups/speech/publi/2011/povey_csl25_elsevier2011_article_p404_439.pdf PDF
- http://www.sciencedirect.com/science?_ob=MImg&_imagekey=B6WCW-50NBNNS-2-F&_cdi=6749&_user=640830&_pii=S088523081000063X&_origin=&_coverDate=04%2F30%2F2011&_sk=999749997&view=c&wchp=dGLbVlz-zSkzS&md5=b7d3a1e0c40cc94760ca24ac5c4ccc71&ie=/sdarticle.pdf PDF
Speech recognition; Gaussian Mixture Model; Subspace Gaussian Mixture Model
Speech recognition based on the Hidden Markov Model-Gaussian Mixture Model (HMM-GMM) framework generally involves training a completely separate GMM in each HMM state.We introduce a model in which the HMM states share a common structure but the means and mixture weights are allowed to vary in a subspace of the full parameter space, controlled by a global mapping from a vector space to the space of GMM parameters.
We describe a new approach to speech recognition, in which all Hidden Markov Model (HMM) states share the same Gaussian Mixture Model (GMM) structure with the same number of Gaussians in each state. The model is defined by vectors associated with each state with a dimension of, say, 50, together with a global mapping from this vector space to the space of parameters of the GMM. This model appears to give better results than a conventional model, and the extra structure offers many new opportunities for modeling innovations while maintaining compatibility with most standard techniques.
@ARTICLE{FITPUB9670, author = "Daniel Povey and Luk\'{a}\v{s} Burget and Mohit Agarwal and Pinar Akyazi and Arnab Ghoshal and Ond\v{r}ej Glembek and K. Nagendra Goel and Martin Karafi\'{a}t and Ariya Rastrow and Richard Rose and Petr Schwarz and Samuel Thomas and et al.", title = "The subspace Gaussian mixture model-A structured model for speech recognition", pages = "404--439", booktitle = "Computer Speech \& Language, Volume 25, Issue 2, April 2011", journal = "Computer Speech and Language", volume = 25, number = 2, year = 2011, publisher = "Elsevier Science", ISSN = "0885-2308", language = "english", url = "https://www.fit.vut.cz/research/publication/9670" }