Videosd

TensorRT accelerated diffusion pipeline, combined with webrtc frontend for camera input and speech recognition.

Github repo: https://github.com/venetanji/videosd

See the full video below: