A self-hosted web app that downloads videos and audio as MP4 or MP3 from 1000+ sites via yt-dlp.
nexmoe/VidBee
Free open-source desktop app to download video and audio from 1000+ sites, then transcribe, search, and summarize them with AI.
Open-source alternative to:4K Video Downloader
Otter.ai
- Stars
- 10,891
- Forks
- 856
- Open issues
- 47
- Last commit
- 2 days ago
What is it?
VidBee is a desktop app for downloading video and audio from 1000+ supported sites such as YouTube, TikTok, Twitter, Instagram, Facebook, Twitch and Bilibili, or importing local media files. Downloads go into a shared queue where you can track progress, pause, resume, retry, and remove tasks.
After a download or import, VidBee can create transcripts with speaker labels and timestamped playback. Speech recognition runs on your computer using local Whisper, SenseVoice, Parakeet, or Qwen3-ASR models. Transcripts are searchable, and you can export them as text or Markdown or burn captions into a new video. AI prompts for summaries, FAQs, translation, and questions run against the transcript through a provider you configure, including OpenAI, Anthropic, Google, Ollama, or LM Studio.
VidBee also supports RSS subscriptions that queue new feed items automatically, with per-subscription keyword filters, directories, and filename templates. It also offers a Docker-based web and API version built on yt-dlp and ffmpeg.
Key features
- 01Downloads video and audio from 1000+ sites, or imports local files.
- 02Local speech recognition with speaker labels and searchable, timestamped transcripts.
- 03AI summaries, translation, and Q&A via OpenAI, Anthropic, Ollama, LM Studio, and other providers.
- 04RSS subscriptions with keyword filters and per-feed download rules.
- 05Control over output formats, metadata, cover art, chapters, and subtitles.
- 06Docker deployment for a web and API version.
Use cases
- Saving videos or podcast episodes from supported sites for offline viewing.
- Turning interviews or lectures into searchable transcripts with speaker labels.
- Automatically downloading new episodes or uploads from RSS feeds.
- Summarizing or translating long recordings with a local or cloud AI model.
Quick start
- 1
Download the desktop app from the download page
- 2
Or run the web and API together locally
pnpm run start:web - 3
Or run with Docker
docker compose up -d --build
Things to know
- AI features send prompts and transcript content to the chosen provider unless you use a local endpoint like Ollama or LM Studio.
- The README says the project is under active development, so expect changes.
- The web/API setup requires Docker or pnpm and a server to run.
Topics
FAQ
- What is VidBee?
- Free open-source desktop app to download video and audio from 1000+ sites, then transcribe, search, and summarize them with AI. VidBee is a desktop app for downloading video and audio from 1000+ supported sites such as YouTube, TikTok, Twitter, Instagram, Facebook, Twitch and Bilibili, or importing local media files. Downloads go into a shared queue where you can track progress, pause, resume, retry, and remove tasks.
- Is VidBee free to use? What license does it have?
- VidBee is open source under the MIT license. Check the license terms before using it in commercial products.
- Is VidBee still maintained?
- Actively maintained. The last commit to VidBee was on Oct 18, 2026.
- What can I use VidBee for?
- Saving videos or podcast episodes from supported sites for offline viewing. Turning interviews or lectures into searchable transcripts with speaker labels. Automatically downloading new episodes or uploads from RSS feeds. Summarizing or translating long recordings with a local or cloud AI model.
- How do I install VidBee?
- Download the desktop app from the download page Or run the web and API together locally: pnpm run start:web Or run with Docker: docker compose up -d --build
- Where can I get VidBee?
- The source code and installation instructions are on GitHub: https://github.com/nexmoe/VidBee
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source, local-first AI dictation app that turns speech into text on macOS and Windows, working offline with Whisper.
Desktop app that transcribes a video with AI and burns styled captions directly into the output video.
Open-source speech-to-text built on Whisper that adds word-level timestamps via forced alignment and optional speaker diarization.
A Claude Code and Agent Skills plugin that lets AI agents watch videos by turning a URL or local file into timestamped frames and a transcript.
Privacy-first AI meeting assistant that records, transcribes and summarizes meetings locally on macOS and Windows, with optional Ollama or cloud LLM summaries.