Download repo:

https://github.com/crowsonkb/v-diffusion-pytorch

Model checkpoints:

https://the-eye.eu/public/AI/models/v-diffusion/cc12m_1_cfg.pth

These are the requirements:

argparse clip k_diffusion sounddevice torch wavfile mtranslate gtts openai-whisper ffmpeg-python mediapipe pygame opencv-python flask-socketio

Excecute:

To execute the api, type flask run inside api folder