Meta’s MoCha AI Creates Animations from Text

Meta, in collaboration with the University of Waterloo, has introduced MoCha, an AI system that generates full character animations from text and speech. Unlike previous models, MoCha creates full-body movements, synchronized lip movements, natural gestures, and interactions between multiple characters. The 30-billion-parameter model produces HD videos of about five seconds and solves lip-syncing issues with a unique mechanism. In tests with 150 scenarios, MoCha outperformed similar models.