DreamTalk is a diffusion-based audio-driven expressive talking head generation framework that can produce high-quality talking head videos across diverse speaking styles. DreamTalk exhibits robust performance with a diverse array of inputs, including songs, speech in multiple languages, noisy audio, and out-of-domain portraits.
- [2023.12] Release inference code and pretrained checkpoint.
conda create -n dreamtalk python=3.7.0
conda activate dreamtalk
pip install -r requirements.txt
conda install pytorch==1.8.0 torchvision==0.9.0 torchaudio==0.8.0 cudatoolkit=11.1 -c pytorch -c conda-forge
conda update ffmpeg
pip install urllib3==1.26.6
pip install transformers==4.28.1
pip install dlib
In light of the social impact, we have ceased public download access to checkpoints. If you want to obtain the checkpoints, please request it by emailing mayf18@mails.tsinghua.edu.cn . It is important to note that sending this email implies your consent to use the provided method solely for academic research purposes.
Put the downloaded checkpoints into checkpoints folder.
Run the script:
python inference_for_demo_video.py \
--wav_path data/audio/acknowledgement_english.m4a \
--style_clip_path data/style_clip/3DMM/M030_front_neutral_level1_001.mat \
--pose_path data/pose/RichardShelby_front_neutral_level1_001.mat \
--image_path data/src_img/uncropped/male_face.png \
--cfg_scale 1.0 \
--max_gen_len 30 \
--output_name acknowledgement_english@M030_front_neutral_level1_001@male_face
wav_path specifies the input audio. The input audio file extensions such as wav, mp3, m4a, and mp4 (video with sound) should all be compatible.
style_clip_path specifies the reference speaking style and pose_path specifies head pose. They are 3DMM parameter sequences extracted from reference videos. You can follow PIRenderer to extract 3DMM parameters from your own videos. Note that the video frame rate should be 25 FPS. Besides, videos used for head pose reference should be first cropped to
image_path specifies the input portrait. Its resolution should be larger than --disable_img_crop like this:
python inference_for_demo_video.py \
--wav_path data/audio/acknowledgement_chinese.m4a \
--style_clip_path data/style_clip/3DMM/M030_front_surprised_level3_001.mat \
--pose_path data/pose/RichardShelby_front_neutral_level1_001.mat \
--image_path data/src_img/cropped/zp1.png \
--disable_img_crop \
--cfg_scale 1.0 \
--max_gen_len 30 \
--output_name acknowledgement_chinese@M030_front_surprised_level3_001@zp1
cfg_scale controls the scale of classifer-free guidance. It can adjust the intensity of speaking styles.
max_gen_len is the maximum video generation duration, measured in seconds. If the input audio exceeds this length, it will be truncated.
The generated video will be named $(output_name).mp4 and put in the output_video folder. Intermediate results, including the cropped portrait, will be in the tmp/$(output_name) folder.
Sample inputs are presented in data folder. Due to copyright issues, we are unable to include the songs we have used in this folder.
If you want to run this program on CPU, please add --device=cpu to the command line arguments. (Thank lukevs for adding CPU support.)
The main goal of this method is to achieve accurate lip-sync and produce vivid expressions across diverse speaking styles. The resolution was not considered in the initial design process. There are two ad-hoc solutions to improve resolution. The first option is to utilize CodeFormer, which can achieve a resolution of
The sample results after super-resolution processing are in the output_video folder.
We extend our heartfelt thanks for the invaluable contributions made by preceding works to the development of DreamTalk. This includes, but is not limited to: PIRenderer ,AVCT ,StyleTalk ,Deep3DFaceRecon_pytorch ,Wav2vec2.0 ,diffusion-point-cloud ,FOMM video preprocessing. We are dedicated to advancing upon these foundational works with the utmost respect for their original contributions.
If you find this codebase useful for your research, please use the following entry.
@article{ma2023dreamtalk,
title={DreamTalk: When Expressive Talking Head Generation Meets Diffusion Probabilistic Models},
author={Ma, Yifeng and Zhang, Shiwei and Wang, Jiayu and Wang, Xiang and Zhang, Yingya and Deng, Zhidong},
journal={arXiv preprint arXiv:2312.09767},
year={2023}
}This method is intended for RESEARCH/NON-COMMERCIAL USE ONLY.