Is this compatible with Hunyuan Avatar?

Hi, Hunyuan Avatar is a new model released for lip sync with audio and image input to generate the person moving and talking. See at https://hunyuanvideo-avatar.github.io/.

Is this repo able to be used with that? Perhaps extract a lora from the existing hunyuan accvideo model? It would absolutely fantastic to get an 8x speedup :)

Thank you!