An alternative project EchoMimic is open-sourced

[lymhust]

EchoMimic is capable of generating portrait videos not only by audios and facial landmarks individually, but also by a combination of both audios and selected facial landmarks.
Project link: https://badtobest.github.io/echomimic.html
GitHub link: https://github.com/BadToBest/EchoMimic

demo_001.mp4

[lhd777]

What a great job this is, providing much inspiration for our Megactor. Everyone is welcome to pay more attention to the aforementioned EchoMimic ☺☺.