Abstract
The article describes a process of creation of a large (800K) vocabulary for a voice-dictation system in Czech language. Such a lexicon has special features different from a lexicon intended for text processing because its look-up entry is a phonetic rather than textual form of a word. It should contain most words and word-forms occurring in standard (non-colloquial) spoken language, it should include separate entries for identical written forms if they are pronounced in different way and, on the other hand, it can omit different spelling variations of the same word. The main goal of this study is to propose and compare various strategies of compiling such a large-scale Czech vocabulary with the aim to keep it as small as possible and at the same time to ensure that it covers the maximum of some in advance unknown text. We demonstrate that we can attain a coverage rate above 97% for a wide spectrum of independent texts when we use a set of about 800,000 most frequent word-forms selected from a 136-million-word corpus.
This work has been supported by the Grant Agency of the Czech Republic (grant no. 102/02/0124) and through research goal project MSM 242200001.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Nouza, J., Drábková, J.: A Dictation System for an 800,000-Word Czech Vocabulary. In: Eurospeech 2003 conference, Geneva (2003) (to appear)
Nejedlová, D.: Building a 20K Vocabulary and Language Model for Czech Language. In: Proc. Speech Processing, 12th Czech-German Workshop. Czech Republic, Prague, pp. 67–70 (2002)
Lingea, S.R.O.: Český korektor pravopisu (c) (1995-1998)
Hajič, J.: The Prague Dependency Treebank. CD-ROM LDC2001T10. Linguistic Data Consortium, Univ. of Pennsylvania (2001)
Nouza, J.: Strategies for Developing a Real-Time Continuous Speech Recognition System for Czech Language. In: Sojka, P., Kopeček, I., Pala, K. (eds.) Proceedings: Text, Speech and Dialogue, pp. 189–196. Springer, Heidelberg (2002)
The Czech National Corpus—SYN2000. Institute of the Czech National Corpus, Faculty of Arts, Charles University, Prague (2000), Available at http://ucnk.ff.cuni.cz
Nejedlová, D., Volejník, M.: Transkripce psaného českého textu do fonetické podoby (Phonetic transcription of printed Czech text). In: Nouza, J (Ed.), Počítačové zpracování řeči. Technical University of Liberec, Liberec, pp. 10–22 (2001)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nejedlová, D., Nouza, J. (2003). Building of a Vocabulary for the Automatic Voice-Dictation System. In: Matoušek, V., Mautner, P. (eds) Text, Speech and Dialogue. TSD 2003. Lecture Notes in Computer Science(), vol 2807. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-39398-6_43
Download citation
DOI: https://doi.org/10.1007/978-3-540-39398-6_43
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-20024-6
Online ISBN: 978-3-540-39398-6
eBook Packages: Springer Book Archive