[mary-users] Creating a Dutch voice with emotions

[Matthijs P]

Dear MARY users,

For a research experiment, I want to create a Dutch TTS voice that can
simulate several emotions. I currently intend to take the following
steps:
1. Create a Dutch TTS voice by following the steps given here
http://mary.opendfki.de/wiki/NewLanguageSupport
2. Use existing filters from 'emofilt' to enable the created voice to
simulate emotions http://emofilt.syntheticspeech.de/index.html
3. Create .wav files for every sentence, in every emotional mode, that
can be used by the software used in the experiment.

I have the following questions about this approach:
1. Looking at the steps given here
http://mary.opendfki.de/wiki/NewLanguageSupport this should be
possible using MARY within a few days, and speaking in the sentences
seems the most work. Is this correct?
2. Do I understand correctly, that existing filters from emofilt can
be used on a new TTS created using MARY?
3. Instead of creating .wav files, could I also apply the filters from
emofilt in real time?
4. Does this approach seem the best way to create a Dutch TTS voice
that can simulate several emotions? If not, do you have suggestions
for a better solution?

Best regards,
Matthijs Pontier