Abstract
In this paper, we propose two methods for speeding up discreteutterance recognition in vocabularies with hundreds to several thousands of words. We show that acceptable results as well as short response time can be achieved if the words are represented by concatenated monophone models (multi-mixture HMMs). In such case, the computation load of the classic Viterbi procedure can be reduced significantly if a proper caching scheme is used. In several experiments done with test vocabularies containing hundreds and thousands of Czech words, we demonstrate that the recognition procedures can be speeded up by a factor of 50 to 100 without a loss of accuracy. The method is well suited for voice controlled systems with a large branching factor and low syntax, i.e. in voice portals, telephone directory assistance, etc.
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
Proc. of Workshop Voice Operated Telecom Services (Do they have a bright future?). Ghent, May 2000.
Nouza J., Holada M.: A Voice-Operated Multi-Domain Telephone Information System. Proc. of 25th Int. Conference on Acoustics, Speech and Signal Processing (ICASSP2000), Istanbul, June 2000 (to appear).
Radová V., Vopálka P.: Methods of Sentences Selection for Read-Speech Corpus Design. Proceedings of the 1st Workshop on Text, Speech and Dialogue, Plzecň 1999, pp. 165–170.
Hájek D.: A Continuous Speech Recognition System. MSc thesis (in Czech). Technical University of Liberec, May 1998.
Hájek D., Nouza J.: A Quasi-Triphone Model Created by Merging Context-Specific Phone Models. In Studientexte zur Sprachkommunikation, Heft 14 (Elektronische Sprachsignalver-arbeitung), Cottbus 1997, pp. 85–92.
Ming J., O’Boyle O., Owens M., Smith F.J.: A Bayesian Approach for Building Triphone Models for Continuous Speech Recognition. IEEE Trans. Speech and Audio Processing, vol. 7, no. 6, 1999, pp. 678–684.
Nouza J., Psutka J., UhlÍř J.: Phonetic Alphabet for Speech Recognition of Czech. Radioengineering, vol. 6, no. 4, Dec 1997, pp. 16–20.
Nouza J., Myslivec M.: Creating and Annotating Speech Database for Continuous Speech Recognition. Proc. of 4th ECMS Workshop, Liberec, May 1999, pp. 147–151.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nouza, J. (2000). A Large Czech Vocabulary Recognition System for Real-Time Applications. In: Sojka, P., Kopeček, I., Pala, K. (eds) Text, Speech and Dialogue. TSD 2000. Lecture Notes in Computer Science(), vol 1902. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45323-7_37
Download citation
DOI: https://doi.org/10.1007/3-540-45323-7_37
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41042-3
Online ISBN: 978-3-540-45323-9
eBook Packages: Springer Book Archive