Abstract
A system for large vocabulary continuous speech recognition of the Slovenian language is described. Two types of modelling units are examined: words and subwords. A data-driven algorithm is used to automatically obtain word decompositions. The performances of one-pass and two-pass decoding strategies were compared. The new models gave promising results. Recognition accuracy was improved by 3.41% absolute at approx. the same recognition time. On the other hand we achieved 30% increase in real time performance at the same recognition error.
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
Kačič, Z., Horvat, B., Zögling, A.: Isues in design and collection of large telephone speech corpus for Slovenian language, LREC 2000.
Young, S., Odell, J., Ollason, D., Kershaw, D., Valtcheva, V., Woodland, P.: The HTK Book, Entropic Inc., 2000.
Zhao, J., Hamaker, J., Deshmukh, N., Ganapathiraju, A., Picone, J.: Fast Recognition Techniques for Large Vocabulary Speech Recognition, Texas Instruments Incorporated, August 15, 1999.
P. Clarkson, R. Rosenfeld: Statistical language modeling using the CMU-Cambridge toolkit. In: Proceedings of EuroSpeech, 1997.
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
Rotovnik, T., Maučec, M.S., Horvat, B., Kačič, Z. (2002). Large Vocabulary Speech Recognition of Slovenian Language Using Data-Driven Morphological Models. 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_46
Download citation
DOI: https://doi.org/10.1007/3-540-46154-X_46
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