If I'm not mistaken, most of them use at&t voices ,
http://www2.research.att.com/~ttsweb/tts/demo.php
granted, thats not their best demo,
part of what separates human voices from synthetic voices is emotion. The notes, and accents we place on words, which computers do not know how to feel on any given statement....for example..
Stop ( or I'll shoot )
Stop (please don't leave )
Stop ( teasing me )
If you want to make a human voice sound robotic, just make all the notes the same , and make all the words equidistant, so as to sound like the person isn't stopping to think / breathe. there is no universal way of saying everything to 'feel ' the same. So if you really want better voice software, you probably want to look for software that allow you to do extensive phoneme editing.
The texture of voice is more important to me, because I could just take any statement and do any adjustments in soundforge or audacity to add more emotions,
Cheers
☯🐉
"To define Tao is to defile it" - Lao Tzu
Edited
13 Years Ago by
planetstardragon