Publication detail

The subspace Gaussian mixture model-A structured model for speech recognition

POVEY, D. BURGET, L. AGARWAL, M. AKYAZI, P. GHOSHAL, A. GLEMBEK, O. GOEL, N. KARAFIÁT, M. RASTROW, A. ROSE, R. SCHWARZ, P. THOMAS, S.

Original Title

The subspace Gaussian mixture model-A structured model for speech recognition

English Title

The subspace Gaussian mixture model-A structured model for speech recognition

Type

journal article in Web of Science

Language

en

Original Abstract

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.

English abstract

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.

Keywords

Speech recognition; Gaussian Mixture Model; Subspace Gaussian Mixture Model

RIV year

2011

Released

01.04.2011

Publisher

Elsevier Science

Location

NEUVEDEN

ISBN

0885-2308

Periodical

COMPUTER SPEECH AND LANGUAGE

Year of study

25

Number

2

State

GB

Pages from

404

Pages to

439

Pages count

36

URL

Documents

BibTex


@article{BUT76383,
  author="Daniel {Povey} and Lukáš {Burget} and Mohit {Agarwal} and Pinar {Akyazi} and Arnab {Ghoshal} and Ondřej {Glembek} and Nagendra {Goel} and Martin {Karafiát} and Ariya {Rastrow} and Richard {Rose} and Petr {Schwarz} and Samuel {Thomas}",
  title="The subspace Gaussian mixture model-A structured model for speech recognition",
  annote="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.",
  address="Elsevier Science",
  booktitle="Computer Speech & Language, Volume 25, Issue 2, April 2011",
  chapter="76383",
  edition="NEUVEDEN",
  howpublished="print",
  institution="Elsevier Science",
  number="2",
  volume="25",
  year="2011",
  month="april",
  pages="404--439",
  publisher="Elsevier Science",
  type="journal article in Web of Science"
}