Voice conversion using duration-embedded bi-HMMs for expressive speech synthesis

• Published in: IEEE transactions on audio, speech, and language processing Vol. 14; no. 4; pp. 1109 - 1116
• Main Authors: Wu, Chung-Hsien, Hsia, Chi-Chun, Liu, Te-Hsien, Wang, Jhing-Fa
• Format: Journal Article
• Language: English
• Published: IEEE 01.07.2006
• Subjects: Algorithm design and analysis; Bi-HMM voice conversion; Computer science; Conversion; Decision trees; embedded duration model; expressive speech synthesis; Hidden Markov models; Humans; Mathematical models; Natural language processing; prosody conversion; Signal synthesis; Spatial databases; Speech; Speech analysis; Speech recognition; Speech synthesis; Testing; Trains; Voice
• ISSN: 1558-7916; 1558-7924
• DOI: 10.1109/TASL.2006.876112

This paper presents an expressive voice conversion model (DeBi-HMM) as the post processing of a text-to-speech (TTS) system for expressive speech synthesis. DeBi-HMM is named for its duration-embedded characteristic of the two HMMs for modeling the source and target speech signals, respectively. Joint estimation of source and target HMMs is exploited for spectrum conversion from neutral to expressive speech. Gamma distribution is embedded as the duration model for each state in source and target HMMs. The expressive style-dependent decision trees achieve prosodic conversion. The STRAIGHT algorithm is adopted for the analysis and synthesis process. A set of small-sized speech databases for each expressive style is designed and collected to train the DeBi-HMM voice conversion models. Several experiments with statistical hypothesis testing are conducted to evaluate the quality of synthetic speech as perceived by human subjects. Compared with previous voice conversion methods, the proposed method exhibits encouraging potential in expressive speech synthesis