Built for Mac · Local-first

Meetings transcribed on your Mac.
Summarized when you want.

Synopsule records your microphone and this Mac's audio, gives you a live transcript while you talk, and — once the recording wraps — labels who said what. Want a summary with action items? That's a button you choose to press. Audio never leaves your Mac.

Runs entirely on this Mac No sign-up to record macOS 15 Sequoia or later

Local by default

Recording and transcription run on this Mac. Raw audio is never uploaded.

Mic + system audio

Captures both sides of remote meetings. No screen recording needed.

Speakers, labeled

Rename a speaker once. Updates everywhere — transcript, search, exports.

Optional summaries

Action items and follow-ups, only when you ask. Off by default.

Features

Built for the way Mac users actually meet.

From the menu bar to a synced playhead, Synopsule fits the muscle memory of macOS. No accounts, no uploads, no surprises.

Whisper, running on your Mac.

Transcripts appear as you talk — fast, accurate, fully offline. Pick a model that matches your Mac: small for speed, large for accuracy. Switch any time.

  • Multiple Whisper model sizes, downloaded once
  • Works offline after first download
  • Tuned for long sessions — an hour or more
  • macOS 15 Sequoia or later, optimized for Apple Silicon
You Microphone
Remote System audio
Local Whisper on device

Speakers, labeled. Renamed once.

When the recording wraps, on-device diarization tells voices apart within each lane — mic and system audio stay separate from the start. Rename "Speaker 2" to a real name and every mention updates — in the transcript, search, and exports.

  • Distinct lanes for microphone and system audio
  • Rename once, propagates everywhere
  • Speaker colors stay consistent
Maya Jordan (remote) Priya Speaker 4
"Speaker 2" Jordan 12 mentions updated
Maya Jordan Priya Speaker 4

Play it back, synced to the transcript.

Click any line to jump there. As audio plays, the active line follows. Skim a 90-minute session in seconds — then dive into the exact moment that matters.

  • Click any line to play that moment
  • Active line follows the playhead
  • Audio stays on your Mac — keep or delete anytime
12:48 / 33:14
MayaEngineering can hit the October window if we trim scope.
MayaLet's lock the launch date so marketing can start the long lead.
JordanI'll send the updated risk register tonight.
PriyaLet me prep customer comms in parallel.

Summaries, action items, follow-ups.

Turn an hour of conversation into a tight summary, a list of owners and tasks, and the open questions to revisit. Strictly opt-in — sending transcript text to AI is a button you choose to press.

  • Concise meeting summary in plain English
  • Action items with owners called out
  • Bring your own key (OpenAI, Anthropic, LiteLLM) — or subscribe

Summary

Locked the launch date for October. Engineering will trim the export redesign to fit; marketing starts long-lead comms immediately.

Action items

  • Jordan — send updated risk register tonight
  • Priya — draft customer comms by Thursday
  • Maya — confirm launch date with leadership

Export anywhere your workflow lives.

Send a transcript to a doc, a video editor, or a project tracker. Speakers, timestamps, and the AI summary travel with the file.

  • Word, PDF, Markdown, plain text, HTML
  • SRT and VTT for video
  • Includes summary, action items, and follow-ups
.docx
Word
.pdf
PDF
.md
Markdown
.html
Web page
.srt
Subtitles
.vtt
Captions
.txt
Plain text
Privacy

Your conversations never leave your Mac.

Recording, transcription, and storage all happen on-device. The one moment data ever moves is if you actively ask for an AI summary — and even then, never the audio.

No accounts. No cloud. No audio uploads.

Synopsule needs no sign-in to record or transcribe. Audio is processed locally — the temporary capture file is deleted after finalization, and the playback copy stays on your Mac so you can replay alongside the transcript. Delete it whenever you like.

Local processing

Whisper runs on this Mac. Recording, transcription, and search never need the network.

Audio stays on this Mac

Raw capture is deleted after the transcript finalizes. The playback copy stays local so you can replay it — keep or delete any time. Audio is never uploaded.

AI is opt-in

Summaries are a deliberate action. Use your own AI key, route through the Synopsule subscription, or never use them at all.

Read the full Privacy Policy.

Pricing

Free to record. Pro for cloud summaries.

Recording, local transcription, speaker labels, playback, and every export format are free, forever. Pay only if you want managed AI summaries.

Free

For everyone, no account needed

$0 forever
  • Local Whisper transcription
  • Mic + system audio capture
  • Speaker labeling & rename
  • Synced playback
  • Every export format
  • Bring your own AI key — direct to provider, no middle service
Get Synopsule

Prefer to pay nothing for AI? Set the app to your own OpenAI, Anthropic, or LiteLLM-compatible key — transcript text goes straight to that provider and never passes through Synopsule.

FAQ

The honest questions, answered.

Where exactly does the audio go?+
Nowhere off this Mac. Audio is captured to a temporary file on your Mac and processed by Whisper running locally. After the transcript finalizes, the temporary capture file is deleted and a playback-ready copy of the audio is kept on your Mac so you can replay alongside the transcript — you can delete it any time. There is no upload, not to Synopsule, not to anyone.
Do I need an account?+
No. Recording, transcription, exports, and search all work without signing in. A subscription is the one optional path that involves an account, and only because the App Store requires it to manage payment.
Does it really work offline?+
Yes — after the first Whisper model download. Pull the cable, fly across an ocean, sit in a basement: Synopsule records and transcribes the same way it does at your desk.
How does it capture the other person on a call?+
Synopsule taps the system audio output the same way macOS itself does — no screen recording, no virtual driver to install. You grant permission once, then it captures both your microphone and whatever is playing through your Mac.
What about AI summaries — does that send my audio away?+
Never the audio. If you ask for a summary, the transcript text (not the audio) is sent to whichever AI provider you've configured: your own OpenAI or Anthropic key directly, or — with a Pro subscription — the Synopsule managed service. If you never summarize, nothing leaves your Mac.
What do I need to run it?+
macOS 15 Sequoia or later, optimized for Apple Silicon (M-series). Pick a smaller Whisper model for snappy transcription on any supported Mac; on M-series the larger models still feel real-time.
How long can a recording be?+
Synopsule is designed for hour-plus sessions. Live transcription is incremental so the UI stays smooth, and exports stream out without loading the whole transcript into memory.
What formats can I export to?+
Word (.docx), PDF, Markdown, plain text, SRT and VTT subtitles. Speaker labels, timestamps, and — if you generated one — the AI summary travel with the file.

Record your next meeting privately.

Synopsule is on the Mac App Store. No sign-up to record, no audio ever leaves your Mac, and the AI is always your choice.