audio-driven-animation

character-policy: Official Benchmark & Codebase for Multimodal 3D Character & Holistic Co-Speech Gesture Policy

Audio-driven talking avatar and lip-sync video generation from a reference image and speech audio.

Few-step diffusion for audio-driven talking head generation making diffusion models speak faster without losing their composure.

Implementing "MakeItTalk" Research paper. It's a project converting an image and an Audio file into a Facial Animated Video Using Reinforcement learning, Convolution Neural Networks(CNNs), Recurrent Neural Networks (RNNs), Generative Adversarial Networks (GANs), LSTM (Long Short-Term Memory) networks

MiniMax H3 video continuation for Wan2GP. Watch the native reframe disappear, then see a 40-second, 3-window result. Overlap replay, bounded video state, real before/after demos, and the full method.

Turn any YouTube creator into an AI clone you can talk to.