Baidu has developed “StyleSync,” software that can perfectly lip-sync (match mouths) any person based on audio data. It can generate highly accurate lip-sync videos from a single sample video.

For example, an English-language version of a movie could be completely reworked in Japanese speech. Since the original language could also be used, the voice actors would lose their jobs.

Demo video for the paper: StyleSync: High-Fidelity Generalized and Personalized Lip Sync in Style-based Generator (CVPR 2023).
https://youtu.be/yAPDl2dVonY

More details on the project page:.
https://hangz-nju-cuhk.github.io/projects/StyleSync