Alibaba just announced FantasyTalking on Hugging Face Realistic Talking Portrait Generation via Coherent Motion Synthesis

discuss with author:

Announcing: Our most advanced speech-to-text model goes beyond accuracy to capture the real-world complexity of human conversation and deliver reliable, source-of-truth audio data. Explore Universal-2 updates 👇

Talking portraits + coherent motion = serious potential for sales avatars, customer support agents and even legacy content revival. Watching this space closely.

Man this looks so realistic 😮

Celebrated too early. I thought the model is on HF😂

That’s ridiculously good

You could literally have Einstein explain relativity himself.

deepfake videos for everyone on your device 👌 but seriously speaking, looks outstanding

@AlibabaGroup GOAT

Looks like regressed tech lol.