Abstract
In the presented work we are dealing with modelling and detection of lexical stress-group (foot) for Czech language. Detection of foot as one type of supra-segmental (prosody) information nearly corresponds to detection of word boundaries. Every native speaker is able to distinguish the feet in continuous speech, but on the other hand there are still no obvious connections between the sound qualities (pitch, intensity, syllable length) and foot prominence realization in Czech. In the experiment we tried to train the Hidden Markov Models (HMM) for Czech feet representation using only pitch information in the syllable nuclei. The most of Czech SPEECON database was used as an experiment source database. A necessary part of the presented system is a tool that transforms given Czech text into the foot units according to the known linguistic rules.
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
Fry, D.B.: Experiments in the perception of stress. Language and Speech 1, 126–152 (1958)
Vicsi, K., Szaszák, G.: Using prosody to improve automatic speech recognition. Speech Commun. 52, 413–426 (2010)
Kroul, M.: Automatic detection of emphasized words for performance enhancement of a czech asr system. In: Proceedings of 13th International Conference Speech and Computer (Specom 2009), Petersburg, Russia, pp. 470–473 (2009)
Kuijk, D.V., van den Heuvel, H., Boves, L.: Using lexical stress in continuous speech recognition for dutch. In: Proc. ICSLP 1996, pp. 1736–1739 (1996)
Volín, J., Zimmermann, J.: Spectral slope parameters and detection of word stress. In: Proceedings of Technical Computing Prague (Humusoft), Praha, pp. 125–129 (2011)
Palková, Z.: Fonetika a fonologie češtiny (Phonetics and phonology of Czech). Karolinum, Praha (1994)
Palková, Z., Volín, J.: The role of f0 contours in determining foot boundaries in czech. In: Proceedings of the 15th ICPhS, Barcelona, vol. 2, pp. 1783–1786 (2003)
Hauser, P.: Základy skladby češtiny. Masarykova Univerzita, Brno (2003)
Bartošek, J.: Czech text-to-foot converter. In: POSTER 2013 (CD-ROM) (2013)
Boersma, P.: Praat, a system for doing phonetics by computer. Glot International 5, 341–345 (2001)
Young, S.: The htk hidden markov model toolkit: Design and philosophy. Entropic Cambridge Research Laboratory, Ltd. 2, 2–44 (1994)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bartošek, J., Hanžl, V. (2013). Foot Detection in Czech Using Pitch Information and HMM. In: Habernal, I., Matoušek, V. (eds) Text, Speech, and Dialogue. TSD 2013. Lecture Notes in Computer Science(), vol 8082. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-40585-3_35
Download citation
DOI: https://doi.org/10.1007/978-3-642-40585-3_35
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-40584-6
Online ISBN: 978-3-642-40585-3
eBook Packages: Computer ScienceComputer Science (R0)