EchoMimicV2: Towards Striking, Simplified, and Semi-Body Human Animation

¹Core Contributor  ²Corresponding Authors

🚀 EchoMimic Series

• EchoMimicV1: Lifelike Audio-Driven Portrait Animations through Editable Landmark Conditioning. GitHub
• EchoMimicV2: Towards Striking, Simplified, and Semi-Body Human Animation. GitHub
• EchoMimicV3: 1.3B Parameters are All You Need for Unified Multi-Modal and Multi-Task Human Animation. GitHub

📣 Updates

• [2025.08.09] 🔥🔥 We update the EchoMimicV3 and release the code.
• [2025.02.27] 🔥 EchoMimicV2 is accepted by CVPR 2025.
• [2025.01.16] 🔥 Please check out the discussions to learn how to start EchoMimicV2.
• [2025.01.16] 🚀🔥 GradioUI for Accelerated EchoMimicV2 is now available.
• [2025.01.03] 🚀🔥 One Minute is All You Need to Generate Video. Accelerated EchoMimicV2 are released. The inference speed can be improved by 9x (from ~7mins/120frames to ~50s/120frames on A100 GPU).
• [2024.12.16] 🔥 RefImg-Pose Alignment Demo is now available, which involves aligning reference image, extracting pose from driving video, and generating video.
• [2024.11.27] 🔥 Installation tutorial is now available. Thanks AiMotionStudio for the contribution.
• [2024.11.22] 🔥 GradioUI is now available. Thanks @gluttony-10 for the contribution.
• [2024.11.22] 🔥 ComfyUI is now available. Thanks @smthemex for the contribution.
• [2024.11.21] 🔥 We release the EMTD dataset list and processing scripts.
• [2024.11.21] 🔥 We release our EchoMimicV2 codes and models.
• [2024.11.15] 🔥 Our paper is in public on arxiv.

🌅 Gallery

Introduction

English Driven Audio

Chinese Driven Audio

⚒️ Automatic Installation

Download the Codes

  git clone https://github.com/antgroup/echomimic_v2
  cd echomimic_v2

Automatic Setup

• CUDA >= 11.7, Python == 3.10

   sh linux_setup.sh

⚒️ Manual Installation

Download the Codes

  git clone https://github.com/antgroup/echomimic_v2
  cd echomimic_v2

Python Environment Setup

• Tested System Environment: Centos 7.2/Ubuntu 22.04, Cuda >= 11.7
• Tested GPUs: A100(80G) / RTX4090D (24G) / V100(16G)
• Tested Python Version: 3.8 / 3.10 / 3.11

Create conda environment (Recommended):

  conda create -n echomimic python=3.10
  conda activate echomimic

Install packages with pip

  pip install pip -U
  pip install torch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1 xformers==0.0.28.post3 --index-url https://download.pytorch.org/whl/cu124
  pip install torchao --index-url https://download.pytorch.org/whl/nightly/cu124
  pip install -r requirements.txt
  pip install --no-deps facenet_pytorch==2.6.0

Download ffmpeg-static

Download and decompress ffmpeg-static, then

export FFMPEG_PATH=/path/to/ffmpeg-4.4-amd64-static

Download pretrained weights

git lfs install
git clone https://huggingface.co/BadToBest/EchoMimicV2 pretrained_weights

The pretrained_weights is organized as follows.

./pretrained_weights/
├── denoising_unet.pth
├── reference_unet.pth
├── motion_module.pth
├── pose_encoder.pth
├── sd-vae-ft-mse
│   └── ...
└── audio_processor
    └── tiny.pt

In which denoising_unet.pth / reference_unet.pth / motion_module.pth / pose_encoder.pth are the main checkpoints of EchoMimic. Other models in this hub can be also downloaded from it's original hub, thanks to their brilliant works:

• sd-vae-ft-mse
• audio_processor(whisper)

Inference on Demo

Run the gradio:

python app.py

Run the python inference script:

python infer.py --config='./configs/prompts/infer.yaml'

Run the python inference script for accelerated version. Make sure to check out the configuration for accelerated inference:

python infer_acc.py --config='./configs/prompts/infer_acc.yaml'

EMTD Dataset

Download dataset:

python ./EMTD_dataset/download.py

Slice dataset:

bash ./EMTD_dataset/slice.sh

Process dataset:

python ./EMTD_dataset/preprocess.py

Make sure to check out the discussions to learn how to start the inference.

📝 Release Plans

⚖️ Disclaimer

This project is intended for academic research, and we explicitly disclaim any responsibility for user-generated content. Users are solely liable for their actions while using the generative model. The project contributors have no legal affiliation with, nor accountability for, users' behaviors. It is imperative to use the generative model responsibly, adhering to both ethical and legal standards.

🙏🏻 Acknowledgements

We would like to thank the contributors to the MimicMotion and Moore-AnimateAnyone repositories, for their open research and exploration.

We are also grateful to CyberHost and Vlogger for their outstanding work in the area of audio-driven human animation.

If we missed any open-source projects or related articles, we would like to complement the acknowledgement of this specific work immediately.

📒 Citation

If you find our work useful for your research, please consider citing the paper :

@article{meng2024echomimicv2,
  title={EchoMimicV2: Towards Striking, Simplified, and Semi-Body Human Animation},
  author={Meng, Rang and Zhang, Xingyu and Li, Yuming and Ma, Chenguang},
  journal={arXiv preprint arXiv:2411.10061},
  year={2024}
}
@article{meng2025echomimicv3,
  title={Echomimicv3: 1.3 b parameters are all you need for unified multi-modal and multi-task human animation},
  author={Meng, Rang and Wang, Yan and Wu, Weipeng and Zheng, Ruobing and Li, Yuming and Ma, Chenguang},
  journal={arXiv preprint arXiv:2507.03905},
  year={2025}
}
@article{meng2026echotorrent,
  title={EchoTorrent: Towards Swift, Sustained, and Streaming Multi-Modal Video Generation},
  author={Meng, Rang and Wu, Weipeng and Yin, Yingjie and Li, Yuming and Ma, Chenguang},
  journal={arXiv preprint arXiv:2602.13669},
  year={2026}
}

🌟 Star History