Abstract
Accurate prediction of prosodic features is one of the critical tasks within a text-to-speech system, especially for under-resourced languages with complex lexical prosody. For synthesized speech to have a natural-sounding intonational contour, an adequate prosodic model should be employed. This study compares the Fujisaki model and the HMM-based prosodic modeling in the context of text-to-speech synthesis, for two quite distant languages with rich prosodic systems: Sesotho, a tonal language from the Bantu family, and Serbian, a South-Slavic language with pitch accent. The results of our experiments suggest that, for both languages, the Fujisaki model outperforms the HMM-based model in the modelling of the intonation contours of utterances of human speech.
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.
Similar content being viewed by others
References
Zerbian, S., Barnard, E.: Word-level prosody in Sotho-Tswana. In: Speech Prosody (2010)
Sečujski, M., Obradović, R., Pekar, D., Jovanov, L., Delić, V.: AlfaNum System for Speech Synthesis in Serbian Language. In: 5th Conf. Text, Speech and Dialogue, pp. 8–16 (2002)
Mixdorff, H.: A novel approach to the fully automatic extraction of Fujisaki model parameters. In: IEEE Int. Conference on Acoustics, Speech, and Signal Processing, Istanbul, Turkey, pp. 1281–1284 (2000)
Mixdorff, H., Fujisaki, H., Chen, G., Hu, Y.: Towards the automatic extraction of Fujisaki model parameters for Mandarin. In: Eurospeech/Interspeech 2003, pp. 873–876 (2003)
Mixdorff, H., Luksaneeyanawin, S., Fujisaki, H.: Perception of tone and vowel quantity in Thai. In: ICSLP (2002)
Gođevac, S.: Transcribing Serbo-Croatian Intonation. In: S.-A. Jun (ed.) Prosodic Typology: The Phonology of Intonation and Phrasing, pp. 146–171. Oxford Linguistics, UK (2005)
Sečujski, M., Jakovljević, N., Pekar, D.: Automatic Prosody Generation for Serbo-Croatian Speech Synthesis Based on Regression Trees. In: Interspeech 2011, pp. 3157–3160 (2011)
Fujisaki, H., Hirose, K.: Analysis of voice fundamental frequency contours for declarative sentences of Japanese. Journal of the Acoustics Society of Japan (E) 5(4), 233–241 (1984)
Taylor, P.A.: The Rise/Fall/Connection model of intonation. In: Speech Communication, 15, 169–186 (1995)
Mixdorff, H., Mehnert, D.: Exploring the Naturalness of Several High-Quality Text-to-Speech Systems. In: Eurospeech 1999, pp. 1859–1862 (1999)
Aguero, P.D., Wimmer, K., Bonafonte, A.: Automatic analysis and synthesis of Fujisaki intonation model for TTS. In: Speech Prosody (2004)
Mixdorff, H., Mohasi, L., Machobane, M., Niesler, T.: A study on the perception of tone and intonation in Sesotho. In: Interspeech 2011, pp. 3181–3184 (2011)
Mohasi, L., Mixdorff, H., Niesler, T.: An acoustic analysis of tone in Sesotho. In: ICPhS XVII, pp. 17–21 (2011)
Dung, T.N., Luong, C.M., Vu, B.K., Mixdorff, H., Ngo, H.H.: Fujisaki model-based F0 contours in Vietnamese TTS. In: ICSLP (2004)
Masuko, T.: HMM-Based Speech Synthesis and Its Applications. Ph.D. thesis, Tokyo Institute of Technology, Japan (2002)
Du Plessis, J.A., et al.: Tweetalige Woordeboek Afrikaans-Suid-Sotho. Via Afrika Bpk, Kaapstad, SA (1974)
Kriel, T.J., van Wyk, E.B.: Pukuntsu Woordeboek Noord Sotho-Afrikaans, Van Schaik, Pretoria, SA (1989)
Khoali, B.T.: A Sesotho Tonal Grammar. PhD Thesis. University of Illinois, Urbana-Champaign, USA (1991)
Sečujski, M., Delić, V.: A Software Tool for Semi-Automatic Part-of-Speech Tagging and Sentence Accentuation in Serbian Language. In: IS-LTC (2006)
Boersma, P.: Praat - A system for doing phonetics by computer. Glot International 5(9/10), 341–345 (2001)
Mixdorff, H.: Intonation Patterns of German - Model-based Quantitative Analysis and Synthesis of F0 Contours. PhD Thesis. TU Dresden, Germany (1998)
Mixdorff, H.: FujiParaEditor (2012), http://public.beuth-hochschule.de/~mixdorff/thesis/fujisaki.html
Pakoci, E., Mak, R.: HMM-based Speech Synthesis for the Serbian Language. In: ETRAN, Zlatibor, Serbia (2012)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Mohasi, L., Sečujski, M., Mak, R., Niesler, T. (2014). A Comparison of Two Prosody Modelling Approaches for Sesotho and Serbian. In: Ronzhin, A., Potapova, R., Delic, V. (eds) Speech and Computer. SPECOM 2014. Lecture Notes in Computer Science(), vol 8773. Springer, Cham. https://doi.org/10.1007/978-3-319-11581-8_4
Download citation
DOI: https://doi.org/10.1007/978-3-319-11581-8_4
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-11580-1
Online ISBN: 978-3-319-11581-8
eBook Packages: Computer ScienceComputer Science (R0)