On-device transcription, zero data retention, and MDM deployment. Steno is built for organizations where privacy isn't optional.
Talk to SalesEvery feature in Steno was designed with a simple question: could a security team approve this? If the answer was no, we didn't ship it.
Offline Mode runs Whisper models locally on Apple Silicon. Audio never leaves the device. No cloud calls, no third-party processors, no network dependency.
Audio is transcribed and immediately discarded. Nothing is written to disk, nothing is sent to our servers. We cannot access, replay, or recover recordings.
Built with Apple's Hardened Runtime protections. Code-signed with Developer ID certificate and notarized through Apple's automated security pipeline.
Microphone activates only when the user holds the hotkey. No background listening, no wake words, no ambient capture. The macOS mic indicator confirms this.
Steno never records, stores, or transmits keystrokes. Accessibility APIs are used solely for text insertion. We have zero visibility into what users type.
When cloud mode is used, audio is sent over TLS directly to the transcription provider. It never touches our infrastructure. Responses are encrypted end-to-end.
Everything your security team will ask about, answered upfront.
No. Audio is never sent to or stored on our infrastructure.
Yes. Offline Mode uses on-device models with zero network access.
Yes. Developer ID signed and notarized by Apple.
Yes. Standard signed PKG. Deploy via Jamf, Intune, Mosyle, Kandji.
No. Mic activates only on explicit user action (hold hotkey).
No. Accessibility is used for text insertion only.
No. We do not use any customer data for model training.
Yes. Standard macOS app removal. No kernel extensions, no daemons.
No agents, no configuration servers, no complex rollouts. Steno is a single signed PKG that works out of the box.
Add the signed, notarized PKG to your MDM console. Jamf, Intune, Mosyle, Kandji — any MDM that supports macOS PKGs.
Assign to device groups or users. Steno installs silently to /Applications and launches automatically. Under 2 MB — installs in seconds.
Hold a key, speak, release. Works in every app immediately. No training required — onboarding takes under 30 seconds.
Voice-to-text saves 30+ minutes per person per day. That's 10 hours per month, per employee.
Agents respond 3x faster with voice. Ticket notes, follow-ups, and internal handoffs — dictated in seconds instead of typed in minutes.
Draft memos, case notes, and contract summaries by voice. Profession-tuned vocabulary recognizes legal terminology out of the box.
CLI-aware mode understands git commands, code terms, and technical jargon. Write docs, PRs, and commit messages without touching the keyboard.
Update CRM notes, draft proposals, and write follow-up emails by voice. Steno adapts tone automatically — casual in Slack, professional in email.
Process documentation, interview notes, onboarding materials — all dictated naturally. 150 words per minute instead of 40.
90+ languages with automatic detection. Team members dictate in their native language. Translate or transliterate on the fly.
Tell us about your team and we'll set up a pilot within 24 hours.