How Listenr works.

Nothing to record. Capture, transcription and speaker recognition run on your Mac and are free forever, as are notes from a local model or your own API key. Pro is $19 a month — or $15 billed yearly, saving 21% — and adds notes from a model we host, Notion delivery and the hosted MCP connector. Team is $39 per user a month, or $29 billed yearly, and is coming soon.

See plans and prices

No. Listenr records the audio your Mac is already playing plus your microphone, so there is no extra participant, no “…has joined the meeting”, and nothing for anyone else to admit or wonder about. As far as the call is concerned, you are just there listening.

No. Audio is captured and transcribed on your Mac, and recordings stay in your local library. If you use an AI model to write the notes, the transcript text is sent to that model — the one you chose — and nothing else is uploaded. Choose a local model and even that stays on your Mac.

No, but you always can. On Pro and Team we run one for you and keep it current, so notes work the moment you sign in. On any plan you can instead point Listenr at a model that runs entirely on your Mac, or at your own Anthropic or OpenAI account.

Google Meet, Zoom and Microsoft Teams are recognised directly: Listenr spots the meeting, offers to record, and reads the speaker names from the call window. Recording itself is not limited to those — anything your Mac plays can be captured, you just may need to label speakers yourself.

Your microphone is you. For everyone else, Listenr watches the meeting window the way you would: whoever the call is highlighting as the active speaker is who it attributes that moment to, and it reads their name from their tile. When the signal is ambiguous it leaves the line unattributed rather than guessing wrong.

Often yes, and the rules genuinely differ by country and by state — some places require everyone on the call to consent. Listenr does not announce itself, so telling people is on you. Check what applies where you and your participants are, and when in doubt, say you are taking notes.

A Mac with Apple Silicon running macOS 15 or later. On first launch Listenr asks for permission to record the screen and system audio, use the microphone, and — if you use Apple’s transcription — recognise speech. It asks for nothing else.

Recording and transcription do, entirely. Writing the notes needs a model: a local one keeps you fully offline, while our hosted model and your own API keys both need a connection. A meeting recorded on a plane transcribes on the plane and can be written up later.

In a Listenr folder inside your Mac’s Application Support directory, readable only by your account. Nothing syncs anywhere unless you share it. Deleting a meeting in the app removes it, and any share link you created for it stops working.

Transcription covers 25 European languages out of the box, with other models available for Chinese, Japanese and Korean, or for very broad coverage. Notes are written in the language of the meeting — Listenr learns which language you use with each person — or in any language you pick.

Yes. Export any meeting as Markdown, send notes straight to Obsidian, Apple Notes or Notion, or share a link that opens in any browser. Your library is plain data on your own disk, so you are never locked in.