Advertisement

International Journal of Speech Technology

, Volume 3, Issue 1, pp 27–34 | Cite as

An Acoustical Study of Syllables of Mandarin Speech

  • Shih-Yu Shen
  • Jesse Wu
  • Hsin-Chuan Lin
Article
  • 55 Downloads

Abstract

In this paper, we study the relations between phoneticfeatures and acoustical signals. Because of the periodicitycharacteristic of voiced sounds, a period of a signal may be expandedin the Fourier series. A hypothesis that thecharacteristics of a simple vowel depend only on a period T andthe coefficients \(\sqrt {a_i^2 + b_i^2 } \) of a period of the simplevowel, where ai and bi are Fourier coefficients, is proposed. The characteristics of a diphthong depend on data of two simplevowels. The hypothesis is verified by synthesis and applied torecognition of vowels. Experiments are done for Mandarin syllables.A rule of changing tones and the characteristic of a stress sound arealso provided in this study.

fourier series phonemes automatic speech recognition 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Chan, S. and Wang, Y. (1995). Tone recognition of continuous Mandarin speech based on neural networks. IEEE Trans. Speech and Audio Processing, 3:146–149.Google Scholar
  2. Committee of Mandarin Teaching Materials of National Taiwan Normal University. (1982). Mandarin Phonetics. Taipei, Taiwan, Cheng Chung Books Co.Google Scholar
  3. Hoemeke, K.A. and Diehl, R.L. (1994). Perception of vowel height: The role of F 1-F 2 distance. J. Acoust. Soc. Am. 96:661–674.Google Scholar
  4. Howie, J.M. (1976). Acoustical Studies of Mandarin Vowels and Tones. London: Cambridge University Press.Google Scholar
  5. Huang, C. (1997). A study of speaker independent recognitions. Master thesis, National Cheng-Kung University, Tainan, Taiwan.Google Scholar
  6. Lee, L., Tseng, C., Gu, H., Liu, F., Chang, C., Chen, H., Lin, Y., Lee, Y., Tu, S., Hsieh, S., and Chen, C. (1993). Golden Mandarin (I)-a real-time Mandarin speech dictation machine for Chinese language with very large vocabulary. IEEE Trans. Speech and Audio Processing, 1:158–179.Google Scholar
  7. Lou, C. (1996). Speaker independent recognitions of final parts for Mandarin. Master thesis, National Cheng-Kung University, Tainan, Taiwan.Google Scholar
  8. Massaro, D.W. and Oden, G.C. (1980). Evaluation and integration of acoustic features in speech perception. J. Acoust. Soc. Am., 67:996–1013.Google Scholar
  9. Ohala, J.J. (1996). Speech perception is hearing sounds, not tongues. J. Acoust. Soc. Am., 99:1719–1725.Google Scholar
  10. O’Shaughnessy, D. (1996). Critique: Speech perception: Acoustic or articulatory? J. Acoust. Soc. Am., 99:1726–1729.Google Scholar
  11. Pitton, J.W., Atlas, L.E., and Loughlin, P.J. (1994). Application of positive time-frequency distributions to speech processing. IEEE Trans. Speech and Audio Processing, 4:554–566.Google Scholar
  12. Rabiner, L.R. and Schafer, R.W. (1978). Digital Processing of Speech Signals. Englewood Cliffs, NJ: Prentice-Hall.Google Scholar
  13. Sheikhzadeh, H. and Dend, L. (1994). Waveform-based speech recognition using hidden filter models: Parameter selection and sensitivity to power normalization. IEEE Trans. Speech and Audio Processing, 2:80–89.Google Scholar
  14. Young, Y. (1995). Speaker independent recognitions of vowels. Master thesis, National Cheng-Kung University, Tainan, Taiwan.Google Scholar
  15. Zhao, Y. (1994). An acoustic-phonetic-based speaker adaptation technique for improving speaker-independent continuous speech recognition. IEEE Trans. Speech and Audio Processing, 2:380–394.Google Scholar

Copyright information

© Kluwer Academic Publishers 1999

Authors and Affiliations

  • Shih-Yu Shen
    • 1
  • Jesse Wu
    • 1
  • Hsin-Chuan Lin
    • 1
  1. 1.Institute of Applied MathematicsNational Cheng-Kung UniversityTainanTaiwan

Personalised recommendations