README.md

metadata

title: EchoMimicV3 Demo
emoji: 🎭
colorFrom: purple
colorTo: blue
sdk: gradio
sdk_version: 4.44.1
app_file: app.py
pinned: false
hardware: a100-large
suggested_storage: large

EchoMimicV3 - Audio-driven Human Animation

Audio/text-driven talking head animation from Ant Group.

Model: BadToBest/EchoMimicV3

Features

• Audio-driven portrait animation
• Text prompt conditioning
• Long video generation with chunking

Requirements

• NVIDIA A100 80GB GPU (required)

Notes

• First generation downloads ~20GB weights (Wan2.1 base + EchoMimic transformer + wav2vec).
• Layout matches the official repo: Wan2.1 base for VAE/T5/CLIP, EchoMimicV3 only for transformer weights.