😭 SadTalker: Learning Realistic 3D Motion Coefficients for Stylized Audio-Driven Single Image Talking Face Animation (CVPR 2023)

Arxiv       Homepage       Github

You may duplicate the space and upgrade to GPU in settings for better performance and faster inference without waiting in the queue.
Alternatively, try our GitHub code on your own GPU.

Possible driving combinations:
1. Audio only 2. Audio/IDLE Mode + Ref Video(pose, blink, pose+blink) 3. IDLE Mode only 4. Ref Video only (all)

Reference Video

How to borrow from reference Video?((fully transfer, aka, video driving mode))

need help? please visit our [best practice page] for more detials
0 45
0 3
face model resolution

use 256/512 model?
preprocess

How to handle input image?
facerender

which face render?
0 10
Examples
Source image Input audio preprocess Still Mode (fewer head motion, works with preprocess `full`) GFPGAN as Face enhancer