From discontinuous to continuous F0 modelling in HMM-based speech synthesis

  title={From discontinuous to continuous F0 modelling in HMM-based speech synthesis},
  author={Kai Yu and Blaise Thomson and Steve J. Young},
The accurate modelling of fundamental frequency, or F0, in HMM-based speech synthesis is a critical factor in achieving high quality speech. However, it is also difficult because F0 values are normally considered to depend on a binary voicing decision such that they are continuous in voiced regions and undefined in unvoiced regions. A widely used solution is to use a multi-space probability distribution HMM (MSDHMM), which directly models discontinuous F0 observations. An alternative solution… CONTINUE READING
9 Citations
12 References
Similar Papers


Publications citing this paper.
Showing 1-9 of 9 extracted citations


Publications referenced by this paper.
Showing 1-10 of 12 references

Simultaneous modelling of phonetic and prosodic parameters, and characteristic conversion for HMM based text-to-speech systems

  • T. Yoshimura
  • Ph.D. dissertation, Nagoya Institute of…
  • 2002
Highly Influential
3 Excerpts

CMU ARCTIC databases for speech synthesis

  • J. Kominek, A. Black
  • Language Technology Institute, School of Computer…
  • 2003
1 Excerpt

A pitch pattern modeling technique using dynamic features on the border of voiced and unvoiced segments

  • H. Zen, K. Tokuda, T. Masuko, T. Kobayashi, T. Kitamura
  • Technical report of IEICE, vol. 101, no. 325, pp…
  • 2001
1 Excerpt

Similar Papers

Loading similar papers…