DSP Robotics Support

i mean like real vocal shots. some kind of lead vocals. can u do this somehow with waveforms or something?

You can synthethise sounds that resemble "oohs" and "aahs" sung by a background choir. For lead vocals you would need samples of entire phrases. However if you repeat them it would be annoying. Our ears are so well trained to the human voice it is impossible to even get close.
Edit: You could try a vocoder.

ah ok good to know.

Yes it's possible but very tricky!

Search for Vocaloid or start with

http://www.yamaha.com/about_yamaha/research/vocaloid/

It would make a truly remarkable Flowstone project, a real showcase, but I wouldn't know where to begin.

I agree with Martin that a vocoder is the easier option and in this way you can make spoken words sing at the pitch you play.

Cheers

Spogg

You can also look for a talented singer, record all kinds of phrases and build a sample library of vowels and consonants. Like Realivox did for Kontakt:
https://www.youtube.com/watch?v=VaVe4LiC0LU

tulamide wrote:You can also look for a talented singer, record all kinds of phrases and build a sample library of vowels and consonants. Like Realivox did for Kontakt:
https://www.youtube.com/watch?v=VaVe4LiC0LU

That's really cool but it strikes me they didn't record and make available enough sounds, so they have more building at a lower level to do. Good though, and I wonder if it can generate random non-words for Jazz scatting...

Cheers

Spogg

i think it use samples.

Spogg wrote:Good though, and I wonder if it can generate random non-words for Jazz scatting...

Yes, you just click together vowels and consonants, not words. So everything is possible.
If words are not so important they even have better fitting libraries:
https://www.youtube.com/watch?v=6WcWFpUbVxo#t=226

Here's an example of an opera singer scatting to jazz (a bit into the video):
https://www.youtube.com/watch?v=zA16RteaOvY

Spogg, you'll love 2:05 and following xD

2:05 lol

yes u need a singer.