AI-Media2Doc

Why we included this project

Most tools that turn podcasts or long videos into text demand an account, a subscription, and handing your media to someone else's servers. AI-Media2Doc takes a different path: it is a self-hosted web app that converts audio or video into written documents in several styles, from Xiaohongshu posts and WeChat articles to study notes, mind maps, and subtitle files. Transcription and rewriting run against AI models you configure, task records stay on your own machine, and there is no login at all. You can also ask follow-up questions about the transcript, and the tool can pull screenshots from the subtitles and insert them into the article without needing a vision model, which keeps the cost near zero. Deployment is a single Docker command, and the browser does the media processing through ffmpeg compiled to WASM, so you never have to install ffmpeg locally.