Publication Details

Dealing with Numbers in Grapheme-Based Speech Recognition

JANDA Miloš, KARAFIÁT Martin and ČERNOCKÝ Jan. Dealing with Numbers in Grapheme-Based Speech Recognition. In: Proceedings of 15th International Conference on Text, Speech and Dialogue. Lecture Notes in Computer Science, 2012, Volume 7499, vol. 2012. Springer-Verlag Berlin Heidelberg 2012: Springer Verlag, 2012, pp. 438-445. ISBN 978-3-642-32789-6. ISSN 0302-9743. Available from: http://www.springerlink.com/content/yx9807202033v381/
Czech title
Rozpoznávání řeči založené na grafémech s důrazem na zpracování číslovek.
Type
conference paper
Language
english
Authors
URL
Keywords

LVCSR, ASR, grapheme, phoneme, speech recognition.

Abstract

Grapheme-based speech recognition approach is suitable in situation of low resource languages, where obtaining of pronunciation dictionary is time- and cost-consuming. The paper describes the process of automatic generation of pronunciation dictionaries with emphasis on the expansion of numbers and presents results on GlobalPhone database.

Annotation

This article presents the results of grapheme-based speech recognition for eight languages. The need for this approach arises in situation of low resource languages, where obtaining a pronunciation dictionary is time- and cost-consuming or impossible. In such scenarios, usage of grapheme dictionaries is the most simplest and straight-forward. The paper describes the process of automatic generation of pronunciation dictionaries with emphasis on the expansion of numbers. Experiments on GlobalPhone database show that grapheme-based systems have results comparable to the phoneme-based ones, especially for phonetic languages.

Published
2012
Pages
438-445
Journal
Lecture Notes in Computer Science, vol. 2012, no. 9, ISSN 0302-9743
Proceedings
Proceedings of 15th International Conference on Text, Speech and Dialogue
Series
Lecture Notes in Computer Science, 2012, Volume 7499
Conference
15th International Conference on Text, Speech and Dialogue, Brno, CZ
ISBN
978-3-642-32789-6
Publisher
Springer Verlag
Place
Springer-Verlag Berlin Heidelberg 2012, DE
DOI
BibTeX
@INPROCEEDINGS{FITPUB10129,
   author = "Milo\v{s} Janda and Martin Karafi\'{a}t and Jan \v{C}ernock\'{y}",
   title = "Dealing with Numbers in Grapheme-Based Speech Recognition",
   pages = "438--445",
   booktitle = "Proceedings of 15th International Conference on Text, Speech and Dialogue",
   series = "Lecture Notes in Computer Science, 2012, Volume 7499",
   journal = "Lecture Notes in Computer Science",
   volume = 2012,
   number = 9,
   year = 2012,
   location = "Springer-Verlag Berlin Heidelberg 2012, DE",
   publisher = "Springer Verlag",
   ISBN = "978-3-642-32789-6",
   ISSN = "0302-9743",
   doi = "10.1007/978-3-642-32790-2\_53",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/10129"
}
Back to top