Faculty of Information Technology, BUT

Publication Details

A Symmetrization of the Subspace Gaussian Mixture Model

POVEY Daniel, KARAFIÁT Martin, GHOSHAL Arnab and SCHWARZ Petr. A Symmetrization of the Subspace Gaussian Mixture Model. In: Proceedings of 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing. Praha: IEEE Signal Processing Society, 2011, pp. 4504-4507. ISBN 978-1-4577-0537-3.
Czech title
Symetrizace Subspace Gaussian Mixture Modelů
Type
conference paper
Language
english
Authors
Povey Daniel (JHU)
Karafiát Martin, Ing., Ph.D. (DCGM FIT BUT)
Ghoshal Arnab (UEDIN)
Schwarz Petr, Ing., Ph.D. (DCGM FIT BUT)
URL
Keywords
Speech Recognition, Hidden Markov Models, Subspace Gaussian Mixture Models
Abstract
We have described a modification to the Subspace Gaussian Mixture Model which we call the Symmetric SGMM. This is a very natural extension which removes an asymmetry in the way the Gaussian mixture weights were previously computed. The extra computation is minimal but the memory used for the acoustic model is nearly doubled. Our experimental results were inconsistent: on one setup we got a large improvement of 1.5% absolute, and on another setup it was much smaller.
Annotation
Last year we introduced the Subspace Gaussian Mixture Model (SGMM), and we demonstrated Word Error Rate improvements on a fairly small-scale task. Here we describe an extension to the SGMM, which we call the symmetric SGMM. It makes the model fully symmetric between the "speech-state vectors" and "speaker vectors" by making the mixture weights depend on the speaker as well as the speech state. We had previously avoided this as it introduces difficulties for efficient likelihood evaluation and parameter estimation, but we have found a way to overcome those difficulties. We find that the symmetric SGMM can give a very worthwhile improvement over the previously described model. We will also describe some larger-scale experiments with the SGMM, and report on progress toward releasing open-source software that supports SGMMs.
Published
2011
Pages
4504-4507
Proceedings
Proceedings of 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing
Conference
International Conference on Acoustics, Speech and Signal Processing 2011, Praha, CZ
ISBN
978-1-4577-0537-3
Publisher
IEEE Signal Processing Society
Place
Praha, CZ
BibTeX
@INPROCEEDINGS{FITPUB9652,
   author = "Daniel Povey and Martin Karafi\'{a}t and Arnab Ghoshal and Petr Schwarz",
   title = "A Symmetrization of the Subspace Gaussian Mixture Model",
   pages = "4504--4507",
   booktitle = "Proceedings of 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing",
   year = 2011,
   location = "Praha, CZ",
   publisher = "IEEE Signal Processing Society",
   ISBN = "978-1-4577-0537-3",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/9652"
}
Back to top