Ancient portraits generating real-person voice-over videos

AI-generated ancient portraits, then dubbing videos of ancient people speaking based on the input audio files.

Deploy Comfyui locally:

Install EchoMimic node: https://github.com/smthemex/ComfyUI_EchoMimic

Download files to folder /models/echo_mimic

https://hf-mirror.com/BadToBest/EchoMimic/tree/main (unet files here)

https://hf-mirror.com/lambdalabs/sd-image-variations-diffusers/tree/main  (other files here)

https://hf-mirror.com/stabilityai/sd-vae-ft-mse/tree/main   (vae file)

Folder structure arranged as follows (where the vae folder location is different from the first version):

├── ComfyUI/models/
|     ├──echo_mimic
|         ├── unet
|             ├── diffusion_pytorch_model.bin
|             ├── config.json
|         ├── audio_processor
|             ├── whisper_tiny.pt
|         ├── vae
|             ├── diffusion_pytorch_model.safetensors
|             ├── config.json
|         ├── denoising_unet.pth
|         ├── face_locator.pth
|         ├── motion_module.pth
|         ├── reference_unet.pth