Description
Voice quality issues have often been studied within the context of formant synthesis in speech synthesis. The NUMBROLA - VOQUAL project is aimed at improving naturalness of concatenative speech synthesis by voice quality analysis(database labeling) and modifications. This project is funded by Région Wallonne, Belgium.
Partners
- TCTS Lab, the Circuit Theory and Signal Processing Lab of the Faculté Polytechnique de Mons
- PERSI, the Situated Perception Lab of the LIMSI-CNRS, Paris