About this project
MyTube provides a browser-based media workflow centered on yt-dlp. Users can add single or multiple URLs, run parallel downloads, track queue progress, upload local videos with generated thumbnails, and organize items into collections. YouTube, Bilibili, and Twitch subscriptions can automatically download new content, while download tracking is intended to prevent duplicate items.
Playback features include a web player with YouTube-style keyboard shortcuts, captions, speed controls, cinema mode, fullscreen, frame stepping, and Up Next navigation. D Mode uses Canvas and WebCodecs for compatible in-car entertainment screens and is intended for passengers rather than drivers. Private RSS links can provide filtered feeds to external readers using channel, author, tag, source, recency, and item-limit filters.
Media management includes optional TMDB metadata matching based on filenames, automatic default subtitle downloads for YouTube and Bilibili, configurable local storage folders for videos, thumbnails, subtitles, avatars, and caches, and optional uploads to OpenList or Alist cloud storage. Advanced download options include cookies.txt uploads for restricted content, custom global yt-dlp arguments, and network proxy configuration.
Administration and integration features include password login, WebAuthn passkeys, an admin gesture login, named read-only visitor accounts, shell task hooks for download events, Telegram notifications, and built-in Cloudflare Tunnel support for remote access without port forwarding. An optional Google Gemini live translation feature can stream the current video's audio to Gemini and play translated speech with live subtitles; the README states the API key is stored server-side and transcripts are not persisted or logged.
The application is built with Node.js, Express, TypeScript, React, Vite, Material-UI, SQLite, and Drizzle ORM. Direct installation requires Node.js 22.22+ or a supported 23 through 26 release, while Docker images and a Compose file are also documented. The ecosystem includes an HTTP API, Chrome extension, native Android client, MikMok short-video client, and an MCP server for AI assistants. The interface supports English, Chinese, Spanish, French, German, Japanese, Korean, Arabic, Portuguese, and Russian.
Comments
0 Rating appears after 10 ratings
Sign in to join the discussion.