Sixth International Conference on Spoken Language Processing
(ICSLP 2000)

Beijing, China
October 16-20, 2000

Prosody Prediction Using a Tree-Structure Similarity Metric

Laurent Blin (1), Mike Edgington (2)

(1) IRISA-ENSSAT, Lannion, France
(2) STAR Lab, SRI International, Menlo Park, USA

In this paper, we present ongoing work on prosody prediction for speech synthesis. Our approach considers sentences as treelike structures and decides on the prosody from a corpus of such structures through tree similarity measurements in a nearest neighbour context. We introduce a syntactic structure and a performance structure representation, the tree similarity metrics considered, and then we discuss the prediction method. Experiments are currently under process to qualify this approach.


Full Paper

Bibliographic reference.  Blin, Laurent / Edgington, Mike (2000): "Prosody prediction using a tree-structure similarity metric", In ICSLP-2000, vol.3, 183-186.