The open-source alternatives to Fireflies.ai are Meetily, OpenWhispr, Buzz and Vibe. Meetily is the closest in shape: it records a live call from your microphone and system audio, transcribes on your machine with Whisper or Parakeet, and writes a summary with a language model you pick, with a local one recommended. OpenWhispr adds call detection and live speaker labels. Buzz and Vibe cover recordings and files with subtitle and document export. All four are MIT. A team switches when call audio cannot leave the building, or when a per-seat notetaker bill stops making sense.
4 of 4 carry an OSI licence. 4 of 4 run on your own machines. Figures read from GitHub on October 26, 2026.
One project per need, with the reason from its own README and the one thing to know before you choose it.
MIT. Records the microphone and system audio together, shows a live transcript from Whisper or Parakeet, and summarises with Ollama locally or a hosted model you choose.
Know firstSpeaker labels, exports to PDF and DOCX, and team features sit in the separate paid PRO edition.
MIT. Detects Zoom, Teams and FaceTime calls, labels speakers on device with voice fingerprints that carry across meetings, and reads Google, Microsoft or Apple calendars.
Know firstLive speaker identification is unavailable on Intel Macs, and team sharing runs through the maintainers' cloud.
MIT. Transcribes and translates files and links offline with Whisper, identifies speakers, and exports TXT, SRT and VTT, with a watch folder and a command line.
Know firstIt records from the microphone; the README does not describe capturing the far side of a call from system audio.
MIT. Files, system audio, the microphone and web links in, with speaker labels and SRT, VTT, TXT, HTML, PDF, JSON and DOCX out. Summaries through Ollama locally or the Claude API.
Know firstThe Claude summary path sends the transcript to Anthropic; the Ollama path keeps it local.
The facts that decide most choices. Each project's page carries the full panel.
| Project | Best for | Openness | Licence | Runs as | Own machines | Stars |
|---|---|---|---|---|---|---|
| Meetily | Live calls, entirely on your machine | OSI | MIT | Desktop app | Yes | 32,187 |
| OpenWhispr | Call detection and speaker labels | OSI | MIT | Desktop app | Yes | 10,041 |
| Buzz | Recordings, files and YouTube links | OSI | MIT | Desktop app, pip, CLI | Yes | 22,202 |
| Vibe | System audio and document exports | OSI | MIT | Desktop app, CLI, Server (HTTP API) | Yes | 7,912 |
Licence and star figures read from GitHub on October 26, 2026. Stars are shown as a dated fact and were not used to rank this page.
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.
MIT. Records the microphone and system audio together, shows a live transcript from Whisper or Parakeet, and summarises with Ollama locally or a hosted model you choose.
Know firstSpeaker labels, exports to PDF and DOCX, and team features sit in the separate paid PRO edition.
A dictation and meeting-notes app under MIT that runs Whisper or Parakeet locally or cloud models with your own key, detects Zoom, Teams and FaceTime calls, and labels speakers on device.
MIT. Detects Zoom, Teams and FaceTime calls, labels speakers on device with voice fingerprints that carry across meetings, and reads Google, Microsoft or Apple calendars.
Know firstLive speaker identification is unavailable on Intel Macs, and team sharing runs through the maintainers' cloud.
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.
MIT. Transcribes and translates files and links offline with Whisper, identifies speakers, and exports TXT, SRT and VTT, with a watch folder and a command line.
Know firstIt records from the microphone; the README does not describe capturing the far side of a call from system audio.
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.
MIT. Files, system audio, the microphone and web links in, with speaker labels and SRT, VTT, TXT, HTML, PDF, JSON and DOCX out. Summaries through Ollama locally or the Claude API.
Know firstThe Claude summary path sends the transcript to Anthropic; the Ollama path keeps it local.
The questions that settle it, in the order they usually come up.
Live: Meetily or OpenWhispr. Recordings and files: Buzz or Vibe.
Meetily (Ollama or a hosted model), Vibe (Ollama or the Claude API), Buzz (a summary plugin), OpenWhispr (the model you configure).
OpenWhispr's team spaces through its hosted service. Meetily's team features are in its paid PRO edition. Buzz and Vibe are single-user.
Fireflies.ai, in the vendor's own words, is a service to "transcribe, summarize, search, and analyze all your team conversations", by its own description. fireflies.ai. The cases below are where that still wins.
Stars, licence, last commit and last release are read from the GitHub API by a script every week and carry the date they were read. Nobody types them.
What each project does is written from its own README and LICENSE, read in full on the date shown, and phrased as the project's claim. The sources are listed at the end.
That a project is faster, better or cheaper than Fireflies.ai. A price for anyone. A user count. A roadmap. If a fact is not on this page, it was not in the source.
Meetily (MIT) is the closest match for live meetings: it captures mic and system audio together, transcribes locally with Whisper or Parakeet, and summarises with Ollama or a hosted model you choose. OpenWhispr (MIT) auto-detects Zoom, Teams and FaceTime calls and labels speakers on device. Buzz and Vibe (MIT) handle recordings and files.
Not among these four. Each records from the machine you are on. OpenWhispr detects when a Zoom, Teams or FaceTime call starts and can read your Google, Microsoft or Apple calendar, which is the nearest thing to automatic capture on this page.
Yes with Meetily, whose README says no data leaves your computer when you use a local model such as Ollama for the summary. Buzz and Vibe transcribe offline; Vibe's summary uses either the Claude API or a local Ollama model. OpenWhispr runs fully local if you choose local models.
Vibe exports SRT, VTT, TXT, HTML, PDF, JSON and DOCX. Buzz exports TXT, SRT and VTT. Meetily's PDF, DOCX and Markdown exports are listed under its PRO edition. OpenWhispr's README does not list export formats.
OpenWhispr has team spaces and web sharing for signed-in users through its hosted service. Meetily's team features are in its paid PRO edition. Buzz and Vibe are single-user desktop apps.
Reveneau sets up the replacement on your own servers, connects it to your product, and builds the software around it.