I just tested the photo talking project memo and it worked pretty well, the project is still in its early stages and I did a lot of fixing of problems when I use it.

A photo, a voice, like a photo talking.

MEMO: Memory-Guided Diffusion for Expressive Talking Video Generation.

comfyui extension: https://github.com/jax-explorer/ComfyUI-IF_MemoAvatar

online run: https://www.comfyonline.app/explore/791e2d71-eaaf-48da-8f99-9b4f7d657ff2