obra/youtube2webpage

This is a script that turns a YouTube video into a webpage you can read. The author finds text easier to learn from than video, which is the stated motivation. You give the script a YouTube URL and a folder name, and it downloads the video, extracts the closed captions, takes screenshots at regular intervals, and assembles everything into a readable HTML page. The output folder contains the webpage itself, a folder of screenshots named by their timestamp in the video, the captions file, a copy of the video, and a stylesheet for the page's appearance. The index.html file is the main readable result, combining the transcript text with the corresponding screenshots so you can follow along visually without watching the video. The script is written in Perl and depends on two external tools that must be installed separately. The first, yt-dlp, handles downloading videos from YouTube. The second, ffmpeg, is a widely used tool for processing video and audio files, used here to extract screenshots. You run the script from the command line by passing a project name and the video URL. There is a live example of the output linked in the README so you can see what a generated page looks like before trying it yourself.