Apple, WHY haven't you stolen the guy who wrote this for ATT?
It's not as simple as that.
AT&T Natural Voices uses actual voice actors, and then breaks their speech down into lots of little bits that it reassembles and tweaks as necessary. It basically punts the problem of synthesizing a convincing human voice (which is not at all to say that it's a straightforward technology, or simple to implement). The amount of work that goes into making a new voice, however, is considerable, and the technology is appropriately expensive.
Vicki is a promising advance in true voice synthesis. Someone just needs to seriously overhaul PlainTalk so that it no longer makes the compromises that were necessary to work with a 16MHz 68020.
Comments
If they could just get rid of those weird "pops" in the voice,
Yup.
Originally posted by Not Unlike Myself
Apple, WHY haven't you stolen the guy who wrote this for ATT?
It's not as simple as that.
AT&T Natural Voices uses actual voice actors, and then breaks their speech down into lots of little bits that it reassembles and tweaks as necessary. It basically punts the problem of synthesizing a convincing human voice (which is not at all to say that it's a straightforward technology, or simple to implement). The amount of work that goes into making a new voice, however, is considerable, and the technology is appropriately expensive.
Vicki is a promising advance in true voice synthesis. Someone just needs to seriously overhaul PlainTalk so that it no longer makes the compromises that were necessary to work with a 16MHz 68020.