Computational Linguistics Colloquium

Thursday, 21 June, 16:15
Seminar Room, Building C7 3
Note: Unusual place!

VocalTractLAb: Towards articulatory text-to-speech synthesis

A system for the articulatory synthesis of speech on the basis of a 3D model of the vocal tract will be presented. Besides the vocal tract model, the system contains a gestural model for the generation of articulatory movements and a module for the aeroacoustic simulation. In contrast to most of the existing articulatory synthesizers, our system is able to synthesize also fricatives and nasals with a high quality. During the talk, also our latest research will be discussed, like the synthesis of singing, the simulation of vocal tract growth, and the possibilities for automatic articulatory text-to-speech synthesis.

If you would like to meet with the speaker, please contact Ingmar Steiner.