This is a CLI to summurize videos into markdown files. It uses whisper.cpp in order to transcirbe the videos locally and then Gemini to generate the markdown from the transcription. It can additionally generates subtitles for your videos, from the whisper output.
This was my first attempt at using AI for coding, I’ve made this project with opencode using Github Copilot with Claude Sonnet 4.1. I still have mixed feelings about it, but it seems a good tool for generating small personal projects.