Meeting transcription

Meetily

A meeting assistant under MIT that records your microphone and system audio together, transcribes live on your machine with Whisper or Parakeet, and summarises with a language model you choose.

OSIAlternative to Otter.ai, Fireflies.ai

What it does

Meetily is "a privacy-first AI meeting assistant that captures, transcribes, and summarizes meetings entirely on your infrastructure", in its README's words. It captures the microphone and the system audio at the same time, which is how it hears both sides of a Zoom, Teams or Meet call, and it shows a live transcript as the meeting happens.

Transcription runs on the machine with Whisper or NVIDIA's Parakeet. Summaries come from a language model you configure: the README recommends Ollama for a local one and also supports Claude, Groq, OpenRouter, OpenAI and any OpenAI-compatible endpoint. It can import an existing recording and transcribe it, a feature marked beta.

What it replaces

Meetily replaces Otter.ai and Fireflies.ai for a person or team that cannot send call audio to a vendor. The README states "All processing is done on your machine. No data ever leaves your computer" when the summary model is local too.

Where it falls short: it records from your machine rather than joining the call as a bot. Speaker labelling, chat with your meetings and calendar integration are listed as coming soon in the separate PRO edition, and PDF, DOCX and Markdown export are listed under PRO. The community edition README does not state its own export formats.

How it runs

A Windows installer (which needs an AVX2-capable CPU) and a macOS .dmg for Apple Silicon. Linux is built from source with pnpm and the project's build script. It is a Tauri app with a Rust backend and a Next.js front end. On macOS, Metal and CoreML are enabled automatically; the Windows build uses Vulkan, and CUDA needs a source build.

The maintainers sell Meetily PRO, "built on a different codebase", with self-hosted deployment, and an Enterprise tier. The README says "Meetily Community Edition will remain free and open source."

Who it fits

A team in a regulated setting, or anyone who wants a transcript and summary of a call with nothing leaving the laptop. A team that wants a bot in every calendar meeting with shared notes should keep a hosted service or look at OpenWhispr's call detection.

Limits the README states

The Windows installer does not select CUDA automatically. Linux requires a source build. Several PRO features are marked coming soon.

Sources

More meeting transcription

ProjectReplacesOpennessStarsLast releaseSelf-host
Buzz

A desktop app under MIT that transcribes and translates audio and video files, YouTube links and a live microphone offline with Whisper, with speaker identification and SRT, VTT and TXT export.

OSI
21,529
Aug 23, 2026
Yes
Vibe

An offline transcription app under MIT that records system audio and the microphone, supports Whisper, Parakeet and Nemotron models, labels speakers, and exports SRT, VTT, TXT, HTML, PDF, JSON and DOCX.

OSI
7,510
Sep 5, 2026
Yes

Added September 16, 2026. Found an error on this page? Write to reveneau@licheo.com and it gets fixed in the next weekly pass.

Running one of these inside your own environment, with your own data and your own security rules, is the kind of work Reveneau does. Read how a forward deployed engagement works.