Run ./main.py to transcribe video(s).
The --all option searches for all files ending in .R.mp4 recursively in the DEFAULT_PATH and transcribes them.
The default model is small, you may choose other models and change the min gap / threshold for speech segments as per your requirements. Refer to whisper and stable-ts documentation for more details.
ffmpeg and stable-ts. Developed on Linux, may or may not work on other platforms.