Skip to content

Publications

Detail of publication

Citation

Jan Vaněk and Zbyněk Zajíc : A Direct Criterion Minimization based fMLLR via Gradient Descend . Text, Speech, and Dialogue, Lecture Notes in Computer Science, vol. 8082, p. 52-59, Springer, 2013.

Download PDF

PDF

Abstract

Adaptation techniques are necessary in automatic speech recognizers to improve a recognition accuracy. Linear Transformation methods (MLLR or fMLLR) are the most favorite in the case of limited available data. The fMLLR is the feature-space transformation. This is the advantage with contrast to MLLR that transforms the entire acoustic model. The classical fMLLR estimation involves maximization of the likelihood criterion based on individual Gaussian components statistic. We proposed an approach which takes into account the overall likelihood of a HMM state. It estimates the transformation to optimize the ML criterion of HMM directly using gradient descent algorithm.

Detail of publication

Title: A Direct Criterion Minimization based fMLLR via Gradient Descend
Author: Jan Vaněk ; Zbyněk Zajíc
Language: English
Date of publication: 1 Sep 2013
Year: 2013
Type of publication: Papers in proceedings of reviewed conferences
Book title: Text, Speech, and Dialogue
Series: Lecture Notes in Computer Science
Číslo vydání: 8082
Page: 52 - 59
DOI: 10.1007/978-3-642-40585-3_8
ISBN: 978-3-642-40584-6
Publisher: Springer
/ 2014-11-12 12:21:25 /

Keywords

ASR, adaptation, fMLLR, gradient descend, Hessian matrix

BibTeX

@INPROCEEDINGS{JanVanek_2013_ADirectCriterion,
 author = {Jan Van\v{e}k and Zbyn\v{e}k Zaj\'{i}c},
 title = {A Direct Criterion Minimization based fMLLR via Gradient Descend},
 year = {2013},
 publisher = {Springer},
 volume = {8082},
 pages = {52-59},
 booktitle = {Text, Speech, and Dialogue},
 series = {Lecture Notes in Computer Science},
 ISBN = {978-3-642-40584-6},
 doi = {10.1007/978-3-642-40585-3_8},
 url = {http://www.kky.zcu.cz/en/publications/JanVanek_2013_ADirectCriterion},
}