An Optimization of Fundamental Frequency and Length of Syllables for Rule-Based Speech Synthesis

  • kyawt Yin Win
  • Tomio Takara
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6485)


In this paper an optimization method has been proposed to minimize the differences of fundamental frequency (F 0) and the differences of length among the speakers and the phonemes. Within tone languages use pitch variation to construct meaning of the words, we need to define the optimized fundamental F 0 and length to obtain the naturalness of synthetic sound. Large variability exists in the F 0 and the length uttered by deferent speakers and different syllables. Hence for speech synthesis normalization of F 0 and lengths are important to discriminate tones. Here, we implement tone rule by using two parameters; optimized F 0 and length. As an advantage in the proposed method, the optimized parameters can be separated to male and female group. The effectiveness of the proposed method is confirmed by the distribution of F 0 and length. Listening tests with high correct rates approve intelligibility of synthetic sound.


Speech Optimization Normalization Myanmar tone Rule-based synthesis 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Lee, M., Lopresti, D.P., Olive, J.P.: A Text-To-Speech Platform for Variable Length Optimal Unit Searching Using Perception Based Cost Function. International Journel Of Speech Technology 6, 347–365 (2003)CrossRefGoogle Scholar
  2. 2.
    Win, K.Y., Takara, T.: Myanmar Speech Synthesis System Using Cepstral Method. In: The International Conference on Electrical Engineering (2008)Google Scholar
  3. 3.
    Win, K.Y., Takara, T.: Rule-based speech synthesis of Myanmar Using Cepstral Method. In: Proceeding of the 11th conference of Oriental-COCOSDA, NICT, Kyoto, Japan, November 25-27, pp. 225–229 (2008)Google Scholar
  4. 4.
    Huang, J., Holt, L.L.: General Perceptual Contributions to Lexical tone normalization. J. Acoust. Soc. Am. 125(6) (June 2009)Google Scholar
  5. 5.
    Zhang, S., Huang, T., Xu, B.: Tone Modeling for Contious Mandarin Speech Recognition. International Journel Of Speech Technology 7, 115–128 (2004)CrossRefGoogle Scholar
  6. 6.
    Myanmar Language Committee, “Myanmar Grammar”, Myanmar Language Committee, Ministry of Education, Myanmar (2005)Google Scholar
  7. 7.
    Thein Tun, U.: Some acoustic properties of tones in Burmese. In: Bradley, D. (ed.) Papers in South – East Asian Linguistics8: Tonation Canberra: Australian National University, pp. 77–116 (1982)Google Scholar
  8. 8.
    Wheatley, J.K.: Burmese. In: Cormier, B. (ed.) The World’s Major Languages, pp. 834–845. Oxford University Press, New YorkGoogle Scholar
  9. 9.
    Do, T.T., Takara, T.: Vietnamese Text-To-Speech system with precise tone generation. Acoust. Sci. & Tech. 25(5), 347–353 (2004)CrossRefGoogle Scholar
  10. 10.
    Noll, A.M.: Cepstrum Pitch Determination. Journal of the Acoustical Society of America 41(2), 293–309 (1967)CrossRefGoogle Scholar
  11. 11.
    Furui, S.: Digital Speech Processing, Synthesis, and Recognition, 2nd edn., pp. 30–31. Marcel Dekker, Inc., New York (2001)Google Scholar
  12. 12.
    Imai, S.: Log Magnitude Approximation (LMA) Filter. Trans. IECE Jpn. J63-A, 886–893 (1980)Google Scholar
  13. 13.
    Xia, Y., Wang, J.: A General Methology for Desiging Globally Convergent Optimization Neural Networks. IEEE Transactions on Neural Networks 9(6) (November 1998)Google Scholar
  14. 14.
    Deng, L., Shaughnessy, D.O.: Speech Processing A dynamic and Optimization-Oriented Approach. Marcel Dekker, Inc., New York (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • kyawt Yin Win
    • 1
  • Tomio Takara
    • 1
  1. 1.Department of Information engineeringUniversity of the RyukyusOkinawaJapan

Personalised recommendations