Privacy policy

Last updated 5 September 2026.

What stays on your device

With the on-device engines (Tiny, Base, Small), your audio is decoded and transcribed inside your browser. It is never uploaded. The speech model is downloaded from Hugging Face’s CDN and cached by your browser. Your last result, your personal terms and your chosen pack are kept in your browser’s local storage.

Cloud engine

If you choose the cloud engine, your audio is compressed in your browser and sent through our server to a speech-recognition provider (Groq or OpenAI) solely to produce the transcript. We do not keep a copy of the audio or the transcript after the response is returned to you.

Corrections you share

When you edit a caption in the studio and “share my text fixes” is on, we receive only the changed words (what we wrote and what you typed) with a few words of surrounding text, the pack, the engine used and your rendering options. We never receive the audio or the full transcript. We use these to improve the packs. Untick the box under the captions to stop sharing at any time.

Payments and minutes

Payments are processed by Stripe; we never see your card details. To keep track of purchased minutes we store, on our database, your checkout email address, the minutes bought and used, the pack and engine of each debit, and a random wallet token that is also kept in your browser’s local storage. Free-minute usage is counted only in your browser.

Analytics and cookies

We do not use advertising trackers. Vercel Analytics records anonymous page views and product events (for example that a file was captioned and which pack was used). Our hosting provider collects standard server logs such as IP address and user agent to operate the service.

Your rights

Email dentaco.alex@gmail.com to ask what we hold about you or to delete it. Clearing your browser storage removes the wallet token, your terms and your last result from that browser.