Publication detail

Phoneme confidence estimator based on Gaussian mixture models

PFEIFER, V. BALÍK, M. MÍČA, I. PRŮŠA, Z.

Original Title

Phoneme confidence estimator based on Gaussian mixture models

Type

journal article - other

Language

English

Original Abstract

The paper propose a statistical approach to phoneme confidence estimation and classification. For this purpose a set of Gaussian Mixture models are trained on the TIMIT corpus.

Keywords

phoneme, classification, models, system

Authors

PFEIFER, V.; BALÍK, M.; MÍČA, I.; PRŮŠA, Z.

RIV year

2009

Released

30. 12. 2009

Publisher

gests

Location

Korea

ISBN

1738-9682

Periodical

International Transactions on Communication and Signal Processing

Year of study

13

Number

13

State

Republic of Korea

Pages from

1

Pages to

8

Pages count

8

BibTex

@article{BUT47792,
  author="Václav {Pfeifer} and Miroslav {Balík} and Ivan {Míča} and Zdeněk {Průša}",
  title="Phoneme confidence estimator based on Gaussian mixture models",
  journal="International Transactions on Communication and Signal Processing",
  year="2009",
  volume="13",
  number="13",
  pages="1--8",
  issn="1738-9682"
}