Voice AI - total beginner

[kidslow]

No, Synthesizer V is based on neural networks and can sound extremely realistic.

Interesting. Will have to take a closer look & listen. Thanks for the clarifying information.

[ztrauq]

If you have a computer with an NVIDIA card and ample VRAM:

https://softology.pro/voc.htm

Machine Learning section of the app can install just about all of the current audio and voice workflows that are open source.

[CinningBao]

Apart from some of the notes at the edge of the model's range, this synthesiserv thing actually does seem to be a level-up from the vocaloid products. The current models seem to be a selection of anime characters though, which I suppose is cheaper than getting an Adele or Aretha model, and is probably more useful if you don't want fake Adele of Aretha.

I wonder which NN model will pick traction more; these 'write your melody and lyrics' engine or the 'input one vocal signal and output another vocal signal'.. I assume people will want the latter more then the former.

[samesung]

If you have a computer with an NVIDIA card and ample VRAM:

https://softology.pro/voc.htm

Machine Learning section of the app can install just about all of the current audio and voice workflows that are open source.

Pretty cool! I haven't got the right GPU yet, but I'm saving the link for later!

[4damind]

Synthesizer V is one of the best AI/vocal synths I've heard so far.