Endless Paradigm

Yes, but, barring the more heavily edited voices, most do still have some little nuances.  Also, rarely will you get a repeat phrase that sounds exactly the same, whereas with Vocaloid, repeat phrases will sound identical.
If to replicate a human voice more, probably better to add some random defects to it - it'll probably make the voice a little more interesting.

Taking away human elements isn't that big of a deal IMO, what matters is the expressability and emotion that differentiates a string of notes to a melody.  This is probably difficult to do in a voice synth.
But then again, a lot of pop songs really just are a string of notes, so it's not that far off.

cute, so cute!! ><'

ZiNgA BuRgA Wrote: [ -> ]what matters is the expressability and emotion that differentiates a string of notes to a melody.  This is probably difficult to do in a voice synth.

Of course this is true. It'll be interesting to see if such a development happens with voice synths (it would be part of trying to make them sound more like a human, I suppose).

Having said that - I'm sure a few of us would agree that some Vocaloid songs have an expressive, emotional level to them. Granted this is elevated by the backing instruments in songs but they do come across quite a bit better then what most would consider a "robot voice" (no emotion or feeling) so in that respect, they are achieving their goal so far.