Abstract
This paper presents a set of’ strategies’ that enabled the development of a real-time continuous speech recognition system for Czech language. The optimization strategies include efficient computation of HMM probability densities, pruning schemes applied to HMM states, words and word hypotheses, a bigram compression technique as well as parallel implementation of the real recognition system. In a series of off-line speaker-independent tests done with 1,600 Czech sentences based on 7,033-word lexicon we got 65%recognition rate. Several on-line tests proved that similar rates can be achieved under real conditions and with response time that is shorter than 1 second.
This work was supported by the Grant Agency of the Czech Republic (grant No. 102/02/0124) and project MSM 242200001. The author wants to thank Tomáš Nouza for his invaluable assistance in multi-thread programming.
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.: A Czech Large Vocabulary Recognition System for Real-Time Applications. In: P. Sojka et al. (Eds.) Text, Speech and Dialogue: Proceedings of the Third International Workshop on Text, Speech, Dialogue. Springer-Verlag, Heidelberg, 2000, pp. 217–222.
Nouza J., olada M.: A Voice-Operated Multi-Domain Telephone Information System. Proc. of 25th Int. Conference on Acoustics, Speech and Signal Processing (ICASSP2000), Istanbul, June 2000, vol.VI, pp. 3755–3758.
Ney H., Ortmanns S.: Dynamic Programming Search for Continuous Speech Recognition IEEE Signal Processing Magazine, Vol. 16, No. 5, Sept. 1999, pp. 64–83.
Nouza J., Psutka J., Uhlíř J.: Phonetic Alphabet for Speech Recognition of Czech. Radioengineering, Vol. 6, No. 4, Dec. 1997, pp. 16–20.
Nejedlová D., Volejník M.: Transkripce psaného českého textu do fonetické podoby (Phonetic transcription of printed Czech text). In: J. Nouza (Ed.), Počítačové zpracování řeči. Technical University of Liberec, 2001, pp. 10–22.
Nejedlová D.: Comparative Study on Bigram Language Models for Spoken Czech Recognition. In: Sojka P. et al. (Eds.): Text, Speech and Dialogue, Proceedings of the Fifth International Conference, Brno, Czech Republic, September 9–12, 2002, pp. 197–204.
Huang X., Acero A., Hon H.-W.: Spoken Language Processing. A Guide to Theory, Algorithm and System Development. Prentice Hall. New Jersey 2001.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nouza, J. (2002). Strategies for Developing a Real-Time Continuous Speech Recognition System for Czech Language. In: Sojka, P., Kopeček, I., Pala, K. (eds) Text, Speech and Dialogue. TSD 2002. Lecture Notes in Computer Science(), vol 2448. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-46154-X_26
Download citation
DOI: https://doi.org/10.1007/3-540-46154-X_26
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-44129-8
Online ISBN: 978-3-540-46154-8
eBook Packages: Springer Book Archive