Privacy

Last updated: 28 October 2026

Who runs this site

This website is run by Joao Jorge, a sole trader based in Portugal, who is the data controller for the personal data described in this policy.
Postal address: Ribeira da Folga – CX Postal 556, Pardieiro, 8550-152 Marmelete, Monchique, Portugal.
Email: [email protected]

Your audio and text are processed in your browser and are not uploaded to us or anyone else.

If you tap Share or Share to WhatsApp, your text is passed to the app you choose. Nothing else is sent. On a computer, Share copies your text instead.

The first time you use it, your browser downloads the speech model (about 80 MB, or about 250 MB if you choose More accurate) from Hugging Face. It stays in your browser's storage on this device until you delete it here or clear this site's data.

Hugging Face hosts the speech model. When your browser downloads it, Hugging Face receives your IP address and browser details, but no audio or text. Hugging Face privacy policy. The download starts only when you first choose or record a voice note.

The microphone is used only while you record, and the recording stays in this tab.

If you choose a speech model under More options, your browser remembers that choice on this device. It is not sent anywhere.

No cookies, no analytics, no ads, no account.

Our host, Cloudflare, sees your IP address when you load the site.

Hosting and legal bases

The site is hosted on Cloudflare Pages, which may process technical data such as IP address for security and delivery. Legal basis: our legitimate interest in running a safe website (Art. 6(1)(f) GDPR).

Transfers

Hugging Face, Inc. (USA) hosts the speech model. When you first choose or record a voice note, your browser downloads the model directly from Hugging Face's servers (huggingface.co and hf.co). As with any website you load, Hugging Face then receives your IP address and browser details; no audio, text or other content is sent. Hugging Face handles this request under its own privacy policy. Hugging Face is not certified under the EU–US Data Privacy Framework (checked 28 October 2026) and its privacy policy does not mention Standard Contractual Clauses; it says data may be processed in the USA. Our legal basis for loading the model from Hugging Face is our legitimate interest in offering a free speech-to-text tool that runs on your device (Art. 6(1)(f) GDPR). If you prefer not to connect to Hugging Face, don't start a voice note: nothing is downloaded from Hugging Face until you do. Cloudflare, Inc. (USA) hosts this site and is certified under the EU–US Data Privacy Framework and its UK Extension, so transfers of technical data to the USA rely on an adequacy decision (Art. 45 GDPR).

Emails and retention

If you email us (see the contact page), we use your message only to reply (legitimate interest, Art. 6(1)(f)) and keep it no longer than 12 months. Hosting logs follow Cloudflare's retention. The speech model and your model choice stay in your browser until you delete them.

Your rights