Impressive

[Inferencer]

Great work, couple of questions which might save a bunch of issues opening up

1. ETA on code release?

2. model released will be 256 or 512?

3. inference time on GPU (please state tested GPU & driving vid frame rate & duration)

4. lowest vram required? (if tested, if not please ignore and await user inputs)

[zhanghongyong123456]

Great work, couple of questions which might save a bunch of issues opening up

1. ETA on code release?
2. model released will be 256 or 512?
3. inference time on GPU (please state tested GPU & driving vid frame rate & duration)
4. lowest vram required? (if tested, if not please ignore and await user inputs)

+1

[lymhust]

An alternative project EchoMimic is open-sourced 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

[Inferencer]

An alternative project EchoMimic is open-sourced 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

to add to that we also now have a SOTA with 512px https://github.com/KwaiVGI/LivePortrait

[mayuelala]

We are going to release the code... don't be worry

[mayuelala]

We release the inference code now！