https://github.com/vgel/repeng is what I was thinking of, but I'm sure there are others.
Alex Young
regularfry
AI & ML interests
None yet
Recent Activity
commented on
an
article
18 days ago
The SOTA Text-to-speech and Zero Shot Voice cloning model that no one knows about...
commented on
an
article
18 days ago
The SOTA Text-to-speech and Zero Shot Voice cloning model that no one knows about...
Organizations
None yet
regularfry's activity
commented on
The SOTA Text-to-speech and Zero Shot Voice cloning model that no one knows about...
18 days ago
commented on
The SOTA Text-to-speech and Zero Shot Voice cloning model that no one knows about...
18 days ago
If it's "just llama", presumably that means you can use existing control vector implementations to steer the model. Would that give you prosody and emotion control independent of the input sample?