ISCA Archive Interspeech 2009
ISCA Archive Interspeech 2009

Voiced/unvoiced decision algorithm for HMM-based speech synthesis

Shiyin Kang, Zhiwei Shuang, Quansheng Duan, Yong Qin, Lianhong Cai

This paper introduces a novel method to improve the U/V decision method in HMM-based speech synthesis. In the conventional method, the U/V decision of each state is independently made, and a state in the middle of a vowel may be decided as unvoiced. In this paper, we propose to utilize the constraints of natural speech to improve the U/V decision inside a unit, such as syllable or phone. We use a GMM-based U/V change time model to select the best U/V change time in one unit, and refine the U/V decision of all states in that unit based on the selected change time. The result of a perceptual evaluation demonstrates that the proposed method can significantly improve the naturalness of the synthetic speech.


doi: 10.21437/Interspeech.2009-138

Cite as: Kang, S., Shuang, Z., Duan, Q., Qin, Y., Cai, L. (2009) Voiced/unvoiced decision algorithm for HMM-based speech synthesis. Proc. Interspeech 2009, 412-415, doi: 10.21437/Interspeech.2009-138

@inproceedings{kang09_interspeech,
  author={Shiyin Kang and Zhiwei Shuang and Quansheng Duan and Yong Qin and Lianhong Cai},
  title={{Voiced/unvoiced decision algorithm for HMM-based speech synthesis}},
  year=2009,
  booktitle={Proc. Interspeech 2009},
  pages={412--415},
  doi={10.21437/Interspeech.2009-138}
}