INTERSPEECH 2007
8th Annual Conference of the International Speech Communication Association

Antwerp, Belgium
August 27-31, 2007

An Automatic Prosody Labeling Method for Mandarin Speech

Chen-Yu Chiang (1), Hsiu-Min Yu (2), Yih-Ru Wang (1), Sin-Horng Chen (1)

(1) National Chiao Tung University, Taiwan
(2) Chung Hua University, Taiwan

A new model-based automatic prosody labeling method for Mandarin speech is proposed. It first introduces four models to describe the relationships of the prosody tags to be labeled, the prosodic features of the speech signals, and the linguistic features of the associated texts. It then employs a sequential optimization procedure to estimate parameters of these four models and find all prosody tags. Experimental results on the Sinica Tree-Bank corpus showed that most prosody tags labeled were meaningful and the estimated parameters of these four models matched well with our a priori knowledge about Mandarin prosody.

Full Paper

Bibliographic reference.  Chiang, Chen-Yu / Yu, Hsiu-Min / Wang, Yih-Ru / Chen, Sin-Horng (2007): "An automatic prosody labeling method for Mandarin speech", In INTERSPEECH-2007, 494-497.