Comments on the Text to Speech "algorithm"

[Gilles Casse]

Christopher Brannon wrote:
> What happens if those WAV files are encoded as 32 kbit/s OGG/Vorbis or
> MP3 files?

Mr Devin experienced latencies with mp3 files. Your suggestions could
improve his solution though.

> I still don't see the appeal of this technique, but to each his own.
> 

This is a 'do it yourself' TTS method: it could helped people to build a
basic speech synthesizer for their own language or dialect without
requiring software skills.

If shared, the valuable data collected would then be available for many
other use.

As an example, I remember a young in early 2000's who was looking for a
TTS in Lakota language for his grandmother.

Best regards,
Gilles