Doug wrote:
The trick would be to do TTS with added data which mapped custom inflection and other notations so the TTS is much closer to what human read audio sounded like. If it was complete enough, one could read the book and annotate it so the TTS knew what to sound like instead of just a flat robotic voice.
I'm confused. How is this any better than just having a human being read and record it?
-- Tracy R Reed http://tracyreed.org -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
