Skip to content

Publications

Detail of publication

Citation

Jan Zelinka and Jan Vaněk and Luděk Müller : Simultaneously Trained NN-based Acoustic Model and NN-based Feature Extractor . Text, Speech, and Dialogue, 18th International Conference, TSD 2015, p. 234-242, 2015.

Abstract

This paper demonstrates how standard feature extraction methods such as PLP can be successfully replaced by a neural network and methods such as mean normalization, variance normalization and delta coefficients can be simultaneously utilized in a neural-network-based acoustic model. Our experiments show that this replacement is significantly beneficial. Moreover, in our experiments, also a neural-networkbased voice activity detector was employed and trained simultaneously with a neural-network-based feature extraction and a neural-networkbased acoustic model. The system performance was evaluated on the British English speech corpus WSJCAM0.

Detail of publication

Title: Simultaneously Trained NN-based Acoustic Model and NN-based Feature Extractor
Author: Jan Zelinka ; Jan Vaněk ; Luděk Müller
Language: Czech
Year: 2015
Type of publication: Book Chapters
Title of journal or book: Text, Speech, and Dialogue, 18th International Conference, TSD 2015
Page: 234 - 242
ISBN: 978-3-319-24032-9
ISSN: 0302-9743
/ 2019-11-18 12:55:41 /

BibTeX

@MISC{JanZelinka_2015_Simultaneously,
 author = {Jan Zelinka and Jan Van\v{e}k and Lud\v{e}k M\"{u}ller},
 title = {Simultaneously Trained NN-based Acoustic Model and NN-based Feature Extractor},
 year = {2015},
 journal = {Text, Speech, and Dialogue, 18th International Conference, TSD 2015},
 pages = {234-242},
 ISBN = { 978-3-319-24032-9},
 ISSN = {0302-9743},
 url = {http://www.kky.zcu.cz/en/publications/JanZelinka_2015_Simultaneously},
}