The goal is to evaluate the quality of the speech produce by an HMM-based synthesis system for French using a new glottal excitation model.
1. For each line of the tab, listen carefully to file1 and file2 (but try to avoid to repeat the playing).
2. Then give a preference score about the timbre of the voice (please don't take the intonation into account), according to the following grades tab:
3. When you answered all questions, verify all questions got an answer
4. Click on the button "Send the answers"
Please, verify that you gave a preference to all questions, then press this button
All recordings are Ircam's property.
Thanks to Gilles Degottex for the php script.