Skip to main content
MIT Licensed • Free • 100% Local

Open Source AI Note Taker for Meetings - 100% Local, MIT Licensed

Free, MIT-licensed AI meeting note taker. Transcription runs on your device, works offline, supports 99+ languages, runs on Windows and macOS. No bots, no subscription required for the Community Edition.

Windows & macOS • 99+ languages • No account required

11.6K+
GitHub stars
180K+
Downloads
1,150+
GitHub forks
100%
Local transcription

What you get, end to end

Six features that matter to anyone searching for an open source AI note taker.

MIT license

Read it, fork it, ship it inside a commercial product. No copyleft, no attribution clause, no field-of-use restriction.

100% local transcription

Audio is transcribed on your device with local Whisper or Parakeet models. Nothing uploaded for the transcript step, ever.

99+ languages

Multilingual transcription out of the box. Mixed-language meetings and accented speech both supported.

Works offline

No internet required for recording or transcription. Useful on flights, in SCIFs, behind air-gapped networks, or when the wifi is just bad.

Self-host or run on your laptop

Runs as a native desktop app on Windows and macOS. For Enterprise teams that want a managed server, contact us for a self-hosted package.

GitHub-first development

Public roadmap, public releases, public issue tracker. 11.6K+ stars, 1,150+ forks, active community contributions.

Open Source AI Note Taker: FAQ

Yes. The Meetily Community Edition is fully open source. The source code lives at github.com/Zackriya-Solutions/meetily, anyone can read it, fork it, audit it, and contribute back. Releases ship as signed Windows and macOS installers built from the same public repo.
MIT License. You can use Meetily commercially, fork it, modify it, and redistribute it. The only requirement is keeping the copyright notice. There is no copyleft, no attribution clause, no field-of-use restriction.
Yes for transcription. Meetily transcribes meetings on your device using local Whisper or Parakeet models, so audio never leaves your machine and you can record without an internet connection. Summaries are pluggable: choose a local model (Ollama or the built-in summarizer) to stay fully offline, bring your own key (BYOK) for Claude, OpenAI, or Groq, or use our Hosted AI option on Pro.
Granola is closed source and syncs meeting data to its cloud by default. Meetily is MIT-licensed open source with 100% local transcription, no cloud sync, and a build-it-yourself path if you want full control. See the detailed feature-by-feature comparison at /vs/granola.
Otter and Fireflies are cloud SaaS products. They upload your audio, transcribe in their data center, and store transcripts on their servers. They also rely on meeting bots that join calls as a separate participant. Meetily records system audio locally with no bot joining, transcribes on your device, and lets you keep summaries local too. Full comparisons at /vs/otter-ai and /vs/fireflies.
The desktop app already runs entirely on your own device, which is the most common self-host shape. For Enterprise deployments that want a managed server (centralized backups, SSO, retention policies), contact meetily@zackriya.com for a self-hosted package.
Yes. Meetily Pro adds Hosted AI summaries, priority support, advanced exports, and team features on top of the open-source Community Edition. Pro is $10/user/month billed annually (regular $25/month). The Community Edition stays free under MIT with no usage caps.
Yes. The repository is github.com/Zackriya-Solutions/meetily. Star it, file issues, or send a pull request. Releases, changelogs, and the public roadmap all live there.
Today: build the app bundle from source via the open-source repo (instructions in the README). A native Linux desktop app is in development; join the waitlist at /granola-for-linux to be notified when it ships. Pro on Linux at launch will be the same tier as macOS and Windows.
We publish a vendor-verified data-retention summary for every cloud LLM Meetily supports, plus the local Ollama path that keeps summaries on-device. Start at /llm-privacy for the index, then read /llm-privacy/anthropic, /llm-privacy/openai, or /llm-privacy/ollama for per-provider zero-data-retention (ZDR), training-default, and retention specifics. Audio never leaves your device regardless of which summary path you pick - only transcript text is sent to a cloud LLM, and only when you choose Hosted AI or BYOK.

Download the open source AI note taker

Join 180K+ users running Meetily on Windows and macOS. MIT licensed, 100% local, free Community Edition.

Windows & macOS • Pro from $10/user/month billed annually