On-device by default
The desktop app runs the speech-to-text model, verse detection, and overlay rendering locally on the operator machine. No audio, transcript, or detection data is uploaded to Kairos servers by default.
Kairos is built around on-device processing. Sermon audio never leaves your machine unless you explicitly opt in to a cloud transcription provider.
Last updated · 2026-06-30
The desktop app runs the speech-to-text model, verse detection, and overlay rendering locally on the operator machine. No audio, transcript, or detection data is uploaded to Kairos servers by default.
If you switch the speech-to-text engine to Deepgram inside the app, your audio stream is sent to Deepgram for transcription under their privacy terms. Kairos acts only as the client; we receive no copy of the stream.
You can revert to local Whisper at any time from Settings → Speech.
If you create an account on the Kairos website (for license management or managed transcription billing), we store your email address, account state, and payment metadata returned by our payment processor (Paystack). We do not store full card details — Paystack handles those under PCI-DSS scope.
Account data is used to operate the service: send receipts, restore licenses, and respond to support requests. It is not sold, rented, or shared for advertising.
The desktop app does not include analytics or crash reporting telemetry. The website uses Vercel Analytics for aggregate pageview counts; no personal identifiers are recorded.
You can request access, correction, or deletion of any account data we hold by emailing nunuolamilekan@gmail.com. We respond within 30 days.
Material changes are noted at the top of this page with a new "Last updated" date. Continued use of Kairos after the change indicates acceptance of the revised policy.