Meta just announced MoCha This AI can create full movie-quality talking & singing characters from just speech & text. 10 wild examples: 1. Talking Characters

2. Emotion Control

3. Action Control

4. Multi-Characters

5. Multi-character Conversation with Turn-based Dialogue

6. Portrait Talking Characters

7.

8.

Project Page: Paper:

If you enjoyed this thread, Follow me @minchoi and please Bookmark, Like, Comment & Repost the first Post below to share with your friends:

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 👇

I'm sorry, this is just not accurate. The lip sync technology is just not ready for prime time YET. The mouths are wonky. Mouths are strangely shaped, open when no words are coming out. This is the area of AI video where the uncanny valley is the highest.

This is the worst it will be