|
Towards articulatory text-to-speech synthesis : Models for the vocal tract, aeroacoustics and control| old_uid | 2558 |
|---|
| title | Towards articulatory text-to-speech synthesis : Models for the vocal tract, aeroacoustics and control |
|---|
| start_date | 2007/03/29 |
|---|
| schedule | 13h30 |
|---|
| online | no |
|---|
| location_info | salle à préciser |
|---|
| summary | Concatenative speech synthesis is currently the leading approach for high quality text-to-speech synthesis, although it has some fundamental limitations regarding its flexibility. Articulatory speech synthesis has obviously a much greater flexibility, but will it replace the concatenative approach in the near future?
In this talk, we present an articulatory speech synthesizer that is being developed in Rostock (Germany) since 2001 and is supposed to produce the best possible speech quality with state-of-the-art models and methods in the field. We introduce the geometric 3D model of the vocal tract, the method for the aeroacoustic simulation and the control concept of the synthesizer. Furthermore, the latest advancements regarding these models and new fields of application will be discussed. |
|---|
| responsibles | Loevenbruck, Welby |
|---|
| |
|