how the synth v 'AI' works

[Bentivi]

I would like to know how this air system works on the synth v.
is she like the Enunu?

I also wanted to know if that's what makes it sound realistic

 

[Halo]

It's similar to ENUNU or Diffsinger (currently a bit closer to diffsinger I'd guess). But it's also just significantly more mature because it's older and a paid software with more than 1 full time developer.

The AI part tries to replicate the dynamics, pronounciation, timing, and pitch of the vocalist; so yes, that's the bit that makes it sound realistic. The vocoder and good voicebank development practices also help, though.

 

[Bentivi]

It's similar to ENUNU or Diffsinger (currently a bit closer to diffsinger I'd guess). But it's also just significantly more mature because it's older and a paid software with more than 1 full time developer.

The AI part tries to replicate the dynamics, pronounciation, timing, and pitch of the vocalist; so yes, that's the bit that makes it sound realistic. The vocoder and good voicebank development practices also help, though.

thank you for the information it will be very useful for me!
do you know if it is possible to make a voice bank by Diffsinger?
sorry if that sounds weird...

 

[Halo]

You can train models for both ENUNU and Diffsinger. You may have to do some work yourself before you can actually do that. Both projects have detailed technical information available via github.