Frequently asked questions

What Stenobar records, where it runs, and what it does with your audio.

What is Stenobar?

Stenobar is a native macOS menu-bar app that records both sides of any call as two separate tracks, their audio and yours, so the transcript always knows who said what. It transcribes on-device, keeps the audio on your disk, and needs no account, server, or subscription. It also summarizes, tags, and searches your recordings, and doubles as a push-to-talk dictation tool.

What are the system requirements?

Stenobar requires macOS 26.0 or later and an Apple Silicon Mac. It is written in Swift and SwiftUI, with no Intel build.

How much does Stenobar cost?

Stenobar is free while it is in beta. With 1.0 a Pro tier arrives as a one-time purchase, no subscription, covering the automation around the recording: automatic meeting capture, calendar-named meetings, the prompt library, and the app-based Thoughts destinations. The recorder, on-device transcription, speaker labels, every export format, your own API keys, dictation, and the full library stay in the free tier. The price will be published before 1.0 ships.

Can Stenobar record Zoom, Teams, or browser audio?

Yes. Stenobar captures everything the system plays, or only the apps you pick, so you can limit a recording to Zoom, Microsoft Teams, Chrome, Safari, or any other single app. Your microphone is captured on its own separate track at the same time.

Does transcription run on my Mac or in the cloud?

Either, and it is your choice. On-device transcription uses Apple Speech, WhisperKit, or Parakeet and never leaves your Mac. If you prefer a cloud engine, you can point Stenobar at AssemblyAI, Deepgram, ElevenLabs, Soniox, OpenAI, Groq, Google Cloud, or Gemini using your own API key. Transcription is entirely optional: with nothing configured, your recordings are still yours to play, export, and keep.

Which AI providers can summarize and tag my recordings?

Summaries and suggested tags can be generated on-device by Apple Intelligence or a local Ollama model, or in the cloud by Anthropic, OpenAI, Google Gemini, OpenRouter, or any OpenAI-compatible endpoint. This is optional and configured separately from transcription, so you can mix providers or skip summaries altogether.

Can Stenobar tell who was speaking?

Yes. A recording that captured both your microphone and system audio gets speaker labels based on the audio itself: your side is labelled "Me" and the other side is labelled separately. This runs entirely on your Mac and works with every transcription provider. You can also give yourself a different name in Settings.

Does Stenobar know when I join a call?

It can. When a meeting app such as Zoom, Microsoft Teams, or a browser running Google Meet takes the microphone, Stenobar offers to record and pre-selects that app as the source. It only checks which apps are using the mic; it never listens in and never accesses the camera. You can switch this to pre-select only, or off, in Settings.

Do my recordings ever leave my Mac?

Audio never leaves your Mac unless you supply your own API key. On-device transcription, speaker labels, and Apple Intelligence never leave the device. Cloud providers are strictly opt-in and only ever touch the single recording you ask them to process, and the app explains what each provider does with your data before you save a key. Your API keys live in the macOS Keychain and nowhere else. Stenobar has no accounts, no sign-in, no analytics, and no telemetry.

What happens to a recording if the app or my Mac crashes?

Every session is written as raw PCM WAV with the header flushed every few seconds, so a crash still leaves usable audio behind.

What audio and transcript formats does Stenobar support?

You can record at 16- or 24-bit, 44.1 or 48 kHz, in mono or stereo. Audio exports to M4A or FLAC, and transcripts export to TXT, Markdown, SRT, or VTT. You can also import existing WAV, MP3, M4A, AIFF, FLAC, or CAF files to transcribe, tag, summarize, and search alongside your own recordings.

Can I combine the system and microphone tracks into one file?

Yes. Stenobar can optionally stitch system audio and your mic into a single mixdown as mono, split stereo, or mixed, and you can re-stitch at any time if you change your mind. The two original tracks stay separate on disk.

How does dictation work?

Hold a hotkey, or just a single modifier key, then speak and release. A floating HUD shows live text, transcribed on-device or streamed through AssemblyAI, Deepgram, ElevenLabs, Soniox, or Gemini, and pasted straight into whatever app you are using. Every take is kept in a Dictation History window you can turn off. A second hotkey captures quick Thoughts and routes them to Apple Reminders, Apple Notes, Todoist, TickTick, Things, Obsidian, Bear, Notion, a Shortcut, or a local inbox.

Can I control Stenobar from Shortcuts or other apps?

Yes. Stenobar ships Shortcuts, Siri, and Spotlight actions to start, stop, or toggle recording, drop a marker, toggle dictation, capture a thought, open the library or Thoughts inbox, and get your last recording. It also registers a stenobar:// URL scheme so tools like Raycast can do the same.

How do I find an old recording?

Stenobar has a three-pane library with full-text search across titles, tags, summaries, and transcript bodies. You can filter by date, sort, star recordings, and group them into projects, each with its own transcription overrides.

Still curious? Read the privacy policy or see what shipped in each release.