Privacy

What we store, and for how long

Your file is uploaded directly from your browser to encrypted object storage. It is never written to SautiScribe's application servers. Once transcription finishes, the audio and the transcript are both deleted automatically within 24 hours. After that the job link stops working.

What we keep alongside it

For each job we record its status, the detected language, the duration, the model used, and the IP address the request came from. The IP address is used only to apply fair-use limits, and it is deleted with the rest of the job.

What we do not do

• We do not use your recordings to train models.
• We do not sell or share your recordings or transcripts with third parties.
• We do not read your transcripts. Nobody at SautiScribe reviews job content.
• We do not use advertising or tracking cookies.

Who processes your audio

Transcription runs on rented GPU infrastructure. Your audio is fetched from storage, processed in memory, and the temporary copy is deleted when the job ends. The speech-recognition model runs on that infrastructure - your audio is not sent to a third-party transcription API.

Recording other people

You are responsible for having the right to upload what you upload. Consent rules for recording conversations differ by country, and in many places all parties must agree.

Deleting something sooner

Because there are no accounts yet, we cannot look a job up on your behalf from an email address alone. If you need something removed before the automatic window elapses, send us the job link. Accounts, transcript history and self-service deletion are planned.