A set of custom nodes integrating OmniVoice, a multilingual zero-shot text-to-speech model from the k2-fsa team, into the image-generation workflow tool ComfyUI has been released. A single model covers more than 600 languages and supports voice cloning from a 3–15 second reference clip as well as multi-speaker dialogue generation.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.