Privacy Policy

1. Who we are

Just Transcribe (“the app”, “we”, “us”) is developed and published by The Dutch Logic Co., a small studio based in the Netherlands. This policy covers two things: the app, and this website at thedutchlogic.nl. They are different in one important way, and we’d rather say it plainly than bury it: the app collects nothing at all, while the website measures visits if, and only if, you say yes to the cookie banner.

For that website measurement, The Dutch Logic Co. is the controller within the meaning of the GDPR. You can reach us by email at support@thedutchlogic.nl, or by post at The Dutch Logic Co., Netherlands. The full contact details are in section 14.

2. How Just Transcribe works

Just Transcribe records audio and converts it into text using speech models that run on your device. They come from more than one place: the Whisper models published by OpenAI and two Distil-Whisper distillations made by third parties, which run via WhisperKit and Apple’s Core ML; NVIDIA’s Parakeet v3 and Canary 1B v2, which run via FluidAudio; and Qwen3-ASR from Alibaba’s Qwen team and Omnilingual ASR from Meta, which run via Apple’s MLX. Summarising, translating and adding punctuation to a finished transcript use a separate local language model. All recording, transcription, summarising, translating, storage and searching happen on your iPhone, iPad, Mac or Apple Watch. This policy also covers the Android app.

• Your audio is not uploaded to us or to any third party.
• Your transcripts are not sent off the device.
• We operate no servers that receive, process or store your recordings or transcripts.
• The app needs the internet for one thing only: downloading the models you choose: a speech model, and the language models used for summaries, punctuation and translation (see section 9). Once they are on the device, recording, transcribing, summarising and translating all work fully offline.

3. Permissions we ask for

The app asks for as little as possible:

• Microphone: so the app can record audio when you tap record. iOS shows the reason as: “Just Transcribe records audio with the microphone so the on-device model can convert your speech into text. Recording can continue while the app is in the background. Nothing is ever sent to the cloud.” Recording only ever starts when you tap record. So that a long recording isn’t cut off when your screen locks or you switch apps, the app can keep recording in the background (iOS shows its recording indicator the whole time, and recording always stops when you tap stop). The app does not use Apple’s Speech Recognition service, so no separate speech-recognition permission is requested.
• Importing a file: if you choose to import an existing audio file, the system file picker lets you select that single file. The app only ever sees files you explicitly pick; it has no general access to your library.
• Local network: only if you open Send to Mac to pair your iPhone with the Mac app. iOS then asks once, so the iPhone can find your own Mac on your own Wi-Fi. Recordings and transcripts go straight between your two devices, encrypted, and never through a server. The iPhone only looks for a Mac you paired; it never makes itself visible on the network. The Mac app only makes itself findable on your network once it is unlocked, and never with your device name.

You can review or revoke the microphone permission at any time under Settings → Just Transcribe, or Settings → Privacy & Security → Microphone. On the Mac and Android apps, the equivalent setting in that system does the same thing.

4. Where your data lives

Recordings, transcripts, the summaries and translations you generate, the models you download and any voices you save are all stored locally in the app’s private storage on your device. You stay in control of them:

• You can delete any recording, transcript, summary or translation inside the app, and remove downloaded models from the model manager. Deleting removes it from the device.
• Saved voices. If you use speaker identification, the app learns a voiceprint for a voice you name and keeps it in the app’s voice library on the device. Voiceprints are created locally, never leave the device, and can be deleted one by one in the app.
• If you have iCloud Backup or device backups enabled, your device’s backup may include the app’s data, exactly as it does for other apps. Those backups are created and protected by Apple under Apple’s Privacy Policy and are controlled entirely by you in your iCloud / device settings. We have no access to them.
• Siri and Spotlight. Off by default. If you switch on Let Siri use my recordings in Settings, the app writes the titles, dates, folders, speakers, summaries and transcripts of your recordings into this device’s own search index, so Siri, Spotlight and Apple Intelligence can find and use them. That index lives on the device, and nothing from it is sent to us. What Siri and Apple Intelligence then do with a request you make is governed by Apple’s Privacy Policy. Switching the setting off removes everything from the index again.
• Uninstalling the app removes its local data from the device.
• Advanced Privacy is a mode that stores nothing at all. A session records into a temporary folder outside the app’s library, erases the audio as soon as the transcript has been read out of it, and deletes the whole folder when you close the screen, when you confirm erasing, or at the next app launch. Nothing from such a session is written to your recordings, and nothing of it can be recovered afterwards. How it works.

5. What the app does not collect

To be completely clear, the app does not:

• require you to create an account or sign in;
• collect your name, email address, phone number or location;
• use analytics, tracking, advertising identifiers or third-party SDKs that profile you;
• track you across other apps or websites (we do not use App Tracking Transparency because there is nothing to track);
• send your recordings, transcripts or usage to us or anyone else;
• sell or share any personal information.

This list is about the app. The website you are reading right now is a different story, and section 6 tells it in full.

6. This website: cookies and Google Analytics

The app and this website are separate things. The app measures nothing. This website does measure visits — but only after you allow it, and you are free not to.

• The first time you open a page on thedutchlogic.nl, a cookie banner appears with two buttons: Accept and Do not accept.
• Until you press Accept, no Google script is loaded at all. Google’s consent mode is set to denied by default for ad_storage, ad_user_data, ad_personalization and analytics_storage, so nothing is stored for advertising or analytics purposes.
• If you press Accept, the site loads Google Analytics 4, measurement ID G-2FXR8863GC, and switches analytics_storage to granted. The three advertising categories stay denied — we run no ads, so there is nothing to store for them.
• If you press Do not accept, nothing is loaded and nothing is measured.
• Your choice is kept in your own browser, in localStorage under the key dl-consent, with the value granted or denied. It stays on your device; the choice itself is never sent to us.
• Withdrawing your consent. Clear this site’s data in your browser — that removes the dl-consent key — and the banner will appear again on your next visit, where you can choose Do not accept. Blocking or deleting cookies for thedutchlogic.nl in your browser settings has the same effect.

What is measured. Standard website statistics: which pages are viewed, roughly which country or region the visit came from, the type of device and browser, and how you arrived. We use it to see which pages actually get read. We do not try to identify individual visitors, we do not build profiles, and we do not combine this with any other data. The measurement data sits in Google’s Analytics property; we keep no copy of it on any server of our own, and how long Google keeps it is set by the data-retention setting in that property. Google’s own handling of this data is described in Google’s privacy policy.

The legal basis for this measurement is your consent (Article 6(1)(a) GDPR, and Article 11.7a of the Dutch Telecommunications Act for storing anything on your device). Without that consent there is no measurement, which is why the banner has a real refuse button rather than only an “OK”.

Apart from Google Analytics after consent, this website loads nothing from third parties: the fonts come from your own system or from our domain, and there are no advertising trackers, social-media buttons or embedded videos.

7. App Store figures and crash reports

The App Store label says “Data Not Collected”, and the first sensible question is: then what does Apple send the developer? Two things, neither of which identifies you:

• App Analytics. Apple provides developers with aggregated, anonymised figures about the app: downloads, sales, active devices, crash counts and similar totals. They come from users who have turned on analytics sharing in iOS, and Apple aggregates them before we ever see them. We cannot trace any of it back to a person, a device, a recording or a transcript.
• Crash reports. For those same users, Apple may pass on crash logs: the technical state of the app at the moment it stopped working. These contain no recordings, no transcripts and no contact details, and we use them only to fix the crash.

You control this in iOS under Settings → Privacy & Security → Analytics & Improvements, where Share iPhone Analytics and Share With App Developers can be switched off. The app itself contains no analytics or crash-reporting SDK — this is Apple’s own system-level reporting, and because what reaches us is aggregated and anonymised by Apple, it is not data collected by the app.

8. Price, unlock & tips

Just Transcribe is free to download, and two speech models stay free forever, so you can record, import and transcribe without paying anything. A single one-time unlock of EUR 9.99 opens everything else, permanently. There is no subscription and no recurring charge. Separately, if you’d like to support development further, the app offers an optional tip (a one-off in-app purchase of EUR 1, EUR 2, EUR 3 or EUR 10), but only if you want to. Tips are entirely voluntary and do not unlock any features.

The unlock and any tip are processed by Apple through the App Store. Your payment is handled under your Apple Account, so we never see or receive your card number, billing address or other payment details. The only thing that comes back to us is the aggregated sales figures described in section 7, which cannot identify you. Apple’s handling of payments is governed by Apple’s Privacy Policy.

9. Third-party services

The app does not embed advertising networks, analytics providers or social media trackers, and there is no third-party SDK that profiles you. These parties are involved only so that the app, or this website, can work:

• Apple: distributes the app, processes the unlock and any optional tips through the App Store, and provides the aggregated figures and crash reports described in section 7.
• The model hosts: when you download a model, the files are fetched over a secure connection from public repositories on Hugging Face: the WhisperKit models hosted by Argmax; Parakeet, Canary and the speaker models converted by Fluid Inference; the MLX community quantisations of Qwen3, Qwen3-ASR and TranslateGemma; and the MLX builds of Omnilingual ASR. This transfers only the model files; none of your recordings, transcripts or personal data are sent. As with any download, a host can see standard technical request data such as your IP address. We do not control those hosts.
• Google: on this website, only after you press Accept, as described in section 6; and on Android as the store the app comes from, as described below. The app for iPhone, iPad, Mac and Apple Watch contains no Google code.

On Android, Google Play takes Apple’s place: it distributes the app and processes the in-app purchases, under Google’s Privacy Policy. We have no other data-sharing arrangements with anyone.

10. Children’s privacy

The app does not collect personal data from anyone, including children. Because nothing is collected, there is nothing we could knowingly gather from a child under 13 (or the equivalent age in your country). The website measurement in section 6 only ever runs after someone accepts the banner, and it is not designed to recognise anyone, child or adult. If you have questions about children’s use of the app, contact us using the details below.

11. Your rights (GDPR / CCPA)

If you are in the European Economic Area, the United Kingdom, California or a similar jurisdiction, privacy laws such as the GDPR and CCPA/CPRA give you rights over your personal data, for example to access, correct or delete it.

Because the app does not collect or hold any personal data about you, we have nothing to access, export, correct or delete on our side, and we never sell or “share” personal information as those laws define it. All of your content stays on your device, where you can manage or delete it yourself at any time. The one processing operation we do carry out is the website measurement in section 6, which runs on your consent, which you can withdraw at any time, and which we cannot link to you as a person.

Complaints. If you believe we handle personal data incorrectly, you have the right to lodge a complaint with a supervisory authority. In the Netherlands that is the Autoriteit Persoonsgegevens (autoriteitpersoonsgegevens.nl). If you live in another EU or EEA country, you may go to the supervisory authority of your own country instead. You are welcome to contact us first — we would rather sort it out directly.

12. Security

Your data is kept in the app’s protected storage and benefits from your device’s built-in encryption and device passcode / Face ID / Touch ID protection. Since your content never travels over a network through us, it isn’t exposed to the risks that come with transmitting and storing data on remote servers. This website is served over HTTPS.

13. Changes to this policy

If the app or this website changes in a way that affects this policy, we’ll update this page and revise the “last updated” date above. If a future version were ever to introduce any data collection in the app, we would tell you clearly and ask for consent where the law requires it.

14. Contact

Questions about privacy, or about anything on this page? Get in touch:

• Email: support@thedutchlogic.nl
• The Dutch Logic Co., Netherlands

The Dutch Logic Co. is the controller for the personal data described in section 6.

← Back to Just Transcribe