8th International Conference on Spoken Language Processing

Jeju Island, Korea
October 4-8, 2004

A Trainable Prosodic Model: Learning the Contours Implementing Communicative Functions within a Superpositional Model of Intonation

Gérard Bailly, Bleicke Holm, Veronique Auberge

CNRS/INPG/ Univ. Stendhal, France

This paper introduces a new model-constrained, data-driven method to generate prosody from metalinguistic information. We refer here to the general ability of intonation to demarcate speech units and convey information about the propositional and interactional functions of these units within the discourse. Our strong hypotheses are that (1) these functions are directly implemented as prototypical prosodic contours that are coextensive to the unit(s) they apply to, (2) the prosody of the message is obtained by superposing and adding all the contributing contours. We describe here an analysis-by-synthesis scheme that consists in identifying these prototypical contours and separating out their contributions in the prosodic contours of the training data. We will show that such a trainable prosodic model generates faithful prosodic contours with very few prototypical movements.

Full Paper

Bibliographic reference.  Bailly, Gérard / Holm, Bleicke / Auberge, Veronique (2004): "A trainable prosodic model: learning the contours implementing communicative functions within a superpositional model of intonation", In INTERSPEECH-2004, 1425-1428.