Private, offline transcription on your own hardware

If your recordings contain anything sensitive — interviews, meetings, medical or legal audio — uploading them to a cloud transcription service means handing that data to a third party. Txtify is open source and self-hosted, so the audio is transcribed on infrastructure you control and never leaves it.

Self-host with Docker See the demo

What "private" actually means here

Be precise about the one exception: translation uses DeepL's API, which is an external network call. Transcription itself is local, but if you enable translation, that text is sent to DeepL. For a fully offline / air-gapped workflow, leave translation off. We'd rather state this plainly than imply Txtify is 100% offline in every configuration.

Why it helps with GDPR and confidentiality

Under GDPR, sending personal data to a third-party transcription service makes that service a processor and adds contractual, transfer and retention obligations. Transcribing on your own infrastructure avoids that hand-off entirely — the data doesn't leave your control. That reduces exposure and simplifies your compliance story.

This is general information, not legal advice. Confirm any setup against your organisation's own requirements and DPA.

Typical private use cases

FAQ

Does anything get uploaded?

Not for transcription — it's local. Only optional DeepL translation sends text off your machine.

Can I run it air-gapped?

Yes, once the Whisper model is downloaded and with translation disabled.

Is it free?

Yes — open source under Apache 2.0. You host it yourself.

Get it on GitHub Compare self-hosted tools

Related: Otter.ai open-source alternative · Best self-hosted transcription tools · Whishper vs Scriberr vs Txtify · Generate SRT subtitles · Transcribe YouTube videos