Privacy Policy
Last updated: November 19, 2026
In short
- Paperharp has no account. The app shows no ads, contains no advertising SDKs, does not track you across other companies’ apps or websites and has no crash reporting.
- Your page photos, imported files, recognised text, books, reading progress and settings stay on your device.
- Text recognition and the device voice run on your device. Google ML Kit, which does part of the text recognition, sends Google usage and performance metrics, never your photos or text.
- Only if you choose the natural voice for a chapter (when available) does that chapter’s text go to our server, which passes it to speech and text-cleanup providers. Our server does not store the text.
- Subscriptions (when available) are paid through the App Store or Google Play and managed with RevenueCat.
- We advertise Paperharp with Apple Ads. On iPhone, Apple’s standard attribution tells us which ad, if any, led to your download, without a tracking prompt and without tracking you across other apps.
- Announcements are delivered with Firebase Cloud Messaging. You can turn them off in the app’s Settings.
This policy explains what happens to your data when you use the Paperharp app on iPhone or Android and when you visit paperharp.com. Paperharp is not yet available in the App Store or Google Play. Features described below as “when available”, such as the natural voice (Premium) and subscriptions, are not offered yet.
1. Who is responsible
The controller of your personal data is Muhammed Tunahan Yollar, an individual developer who publishes the app under the brand “Paperharp” (“we”, “us”). We are based in Türkiye.
For any privacy question or request, write to info@paperharp.com.
2. What stays on your device
Paperharp does not ask you to create an account. The following are stored only on your device, inside the app’s own storage:
- photos of book pages you take or choose, and cover images;
- files you import (PDF, EPUB, Word and text files);
- the text recognised from your pages;
- your books, chapters, bookmarks and reading progress;
- audio files of chapters prepared with the natural voice;
- your settings, and small counters the app uses on the device (for example, to decide when to show the store rating prompt).
We do not receive any of this. Your page photos are not sent to us or to our providers.
Processing on your device
- Text recognition (OCR). On iPhone, Paperharp uses Apple’s Vision framework, and Google ML Kit for pages in Hindi (Devanagari script). On Android, it uses Google ML Kit for all text recognition; the recognition model is downloaded through Google Play services, which operate under Google’s own terms. Your images and the recognised text are processed on the device and are not sent to Google.
- Language detection for chapters and sentences also runs on the device: with Apple’s NaturalLanguage framework on iPhone and Google ML Kit Language ID on Android.
- Google ML Kit usage metrics. When ML Kit is used, it sends Google usage and performance metrics: device information (manufacturer, model, operating system version), app information (bundle ID, app version), a per-installation identifier that Google says is not intended to identify a user or a device, performance metrics (such as latency), API configuration (such as image format and resolution), event types and error codes, and, for language identification, the languages identified. Google uses them for diagnostics, usage analytics, maintaining and improving its APIs and detecting abuse, and processes them as its own under the ML Kit terms. They do not include your photos or text. Google describes this data for iOS and Android.
- Device voice. Reading aloud with the device voice uses your phone’s built-in speech engine on the device. One exception: on Android, if you pick a system voice that needs a network connection, your operating system may send the text to its speech provider (for example Google). Voices that work offline avoid this.
- “Preparation finished” notifications are created on your device and are not sent through our server.
Device backups
The app’s data can be included in your operating system’s device backup (iCloud Backup on iPhone, Android backup on Android). Those backups are handled by Apple or Google under your account settings with them, not by us.
Things you share yourself
- Share cards (a quote image) are created on your device. They leave your device only if you share them, and only to the place you choose in your system’s share sheet.
- The store rating prompt belongs to your operating system (Apple App Store or Google Play). Any rating or review you leave is handled by Apple or Google.
3. Natural voice (Premium, when available)
The natural voice is an optional feature you choose per chapter. Chapters read with the device voice never use it. When you prepare a chapter with the natural voice:
What is sent to our server
- the text of that chapter, with its language and the voice you chose;
- a random installation ID created by the app on your device, and a user ID: the anonymous RevenueCat app user ID (see section 5), or the installation ID when RevenueCat is not in use (neither is linked to your name, email or any account);
- your IP address, which any internet connection reveals.
When you open the natural voice settings, the app also asks our server for the list of available voices. That request contains no book text.
What our server does
- It passes the text to one of our speech providers to turn it into audio: Google Cloud Text-to-Speech, Microsoft Azure Speech or ElevenLabs. To make the reading sound natural, the speech provider may also receive the neighbouring sentences of the same chapter.
- It passes the text to Google Gemini to tidy up the recognised text before it is read aloud.
- It returns the audio to your device, where it is saved with the chapter so you can listen offline.
- It uses the user ID to check with RevenueCat whether your subscription is active, and to count the characters voiced for your monthly allowance.
Our server does not store the text. It does not save it, keep a copy of it or write it to its logs; the text is held in memory only while your request is being processed. Our server’s logs contain technical information only, such as the time, status and duration of a request and character counts.
Your IP address and installation ID are kept in memory for about 60 seconds, only to prevent abuse and to enforce rate limits, and are then discarded.
To apply your monthly allowance, our server stores, for each user and calendar month (UTC), the number of characters voiced with the natural voice. This counter is stored under a SHA-256 hash of the user ID, never the ID itself, contains no text, and is deleted after 13 months.
The providers listed above process the text under their own terms and may store it for their own purposes, for example to monitor abuse. Their privacy policies are linked in section 8. If you prefer that your text does not leave your device, use the device voice.
4. Notifications
- To deliver news and announcements, the app uses Firebase Cloud Messaging (Google), which creates a device token. On iPhone, notifications are delivered through Apple’s Push Notification service.
- The app subscribes to general announcement topics: one for all users and one for your app language. This subscription can happen before you grant notification permission.
- We send announcements to these topics, not to individual devices, and we do not collect device tokens on our server.
- You can turn announcements off with the “News and announcements” switch in the app’s Settings. This unsubscribes the app from all announcement topics. The Firebase software in the app may still keep a device token.
- Notifications are only shown if you allow them in your operating system. You can change this at any time in your system settings.
5. Subscriptions (when available) and ad measurement
When subscriptions become available, they will be processed by the Apple App Store or Google Play, under their own terms and privacy policies. Your payment details are not shared with us. You renew, change and cancel your subscription in your store account, and refunds are handled under the store’s rules.
RevenueCat
To manage subscriptions on iPhone and Android, we use RevenueCat (RevenueCat, Inc., USA). The RevenueCat software in the app creates an anonymous app user ID and sends RevenueCat purchase and receipt data, your subscription status, device and app information and your IP address, so that purchases can be validated and your subscription managed. The app sends this RevenueCat app user ID (or, without RevenueCat, the random installation ID) to our server as your user ID (see section 3). RevenueCat processes this data on our behalf; see RevenueCat’s privacy policy.
Apple Ads attribution (iPhone)
We advertise Paperharp with Apple Ads in the App Store. Paperharp itself shows no ads. On iPhone (iOS 14.3 or later), the RevenueCat software collects an attribution token from Apple’s AdServices framework and uses it to ask Apple which Apple Ads campaign, ad group and keyword, if any, led to your download. This is Apple’s standard attribution: it does not use the App Tracking Transparency permission, and it is not used to track you across other companies’ apps or websites. We use the result to see which ads lead to downloads and subscriptions (for example, the cost per subscriber). Apple provides this attribution under its own terms (see section 8).
6. This website and email
- paperharp.com uses no cookies, no analytics and no third-party scripts or fonts. Like any web server, ours receives technical request data (such as your IP address, browser type and the page requested) and may record it in server logs for operation and security. The app opens this website when you view our legal pages from Settings.
- Email. If you write to us, we receive your email address and whatever you include in your message, and use it to reply to you.
7. Legal bases
Under the EU General Data Protection Regulation (GDPR), we rely on the following legal bases:
- Performance of a contract (Art. 6(1)(b)): to provide the features you use, including preparing chapters with the natural voice, subscriptions when available (managed with RevenueCat, including checking your subscription status and counting your monthly allowance), and answering your support requests.
- Legitimate interests (Art. 6(1)(f)): to keep our server secure, prevent abuse and apply rate limits (IP address, installation ID, technical logs); to deliver general app announcements through topics (device token and topic subscriptions); to measure which Apple Ads campaigns lead to downloads and subscriptions (Apple Ads attribution on iPhone); and to use Google ML Kit for on-device text recognition and language identification, which sends Google the usage and performance metrics described in section 2. You can object at any time, for example by turning off “News and announcements”.
- Consent (Art. 6(1)(a)): showing notifications on your device, which you allow through your operating system’s notification permission. You can withdraw it at any time in your system settings.
- Legal obligation (Art. 6(1)(c)): where we must keep or process data to comply with the law, for example when answering a request to exercise your rights.
8. Processors and recipients
We use the following service providers. They process data on our behalf for the purpose shown:
| Provider | Purpose | Location | Privacy |
|---|---|---|---|
| Hetzner Online | Hosting our server and this website | Germany | Privacy policy |
| Google Cloud Text-to-Speech | Natural voice: turning text into speech | Google’s global infrastructure, which may include the USA | Privacy notice |
| Microsoft Azure Speech | Natural voice: turning text into speech | EU (West Europe region) | Privacy statement |
| ElevenLabs | Natural voice: turning text into speech | USA | Privacy policy |
| Google Gemini API | Natural voice: tidying recognised text | Google’s global infrastructure, which may include the USA | Terms · Privacy policy |
| RevenueCat | Managing subscriptions: validating purchases, subscription status | USA | Privacy policy |
| Firebase Cloud Messaging (Google) | Delivering announcements | Google’s global infrastructure, which may include the USA | Privacy and security |
| Apple Push Notification service (iPhone only) | Delivering notifications to iPhone | Apple’s global infrastructure, which may include the USA | Privacy policy |
The following services are run by Apple or Google under their own terms, as independent controllers, when you use them through your device: the App Store and Google Play (app download, subscriptions when available, ratings), Google Play services (downloading the text recognition model on Android), your system’s speech provider (only for network voices on Android) and iCloud or Android backups. Two more are run by Google and Apple as independent controllers under their own terms: Google ML Kit, for the usage and performance metrics it sends Google (section 2; ML Kit terms), and Apple Ads attribution through Apple’s AdServices framework on iPhone (section 5).
We do not sell your personal data, and we do not share it with advertisers or data brokers. We may disclose data to authorities where the law requires it.
9. International transfers
Our server is hosted in Germany, in the European Union. Some of our providers may process data outside the European Economic Area and Türkiye, in particular in the USA (see the table above), including RevenueCat, Inc., which manages subscriptions from the USA. Where this happens, the transfer relies on the safeguards the provider offers, such as an adequacy decision of the European Commission (for example the EU–US Data Privacy Framework, for certified companies) or the European Commission’s Standard Contractual Clauses in the provider’s data processing terms. You can ask us for more information about these safeguards at info@paperharp.com.
If you are in Türkiye, see also our KVKK information notice (Aydınlatma Metni).
10. How long data is kept
- Data on your device: until you delete the book or chapter, or delete the app. Copies in device backups follow your backup settings.
- Natural voice text on our server: not stored; held in memory only while the request is processed.
- IP address and installation ID on our server: about 60 seconds, in memory.
- Monthly natural-voice counters on our server (characters per user and calendar month, under a SHA-256 hash of the user ID, without text): deleted after 13 months.
- Technical server logs (without text): for as long as needed to operate and secure the service.
- Data held by our providers, including subscription data at RevenueCat: under each provider’s own terms (see section 8). Google ML Kit metrics and Apple Ads attribution data are kept by Google and Apple under their own terms.
- Announcement topics: until you turn off “News and announcements” or delete the app. Device tokens are kept by Firebase under Google’s terms.
- Emails: as long as needed to handle your request and any follow-up.
11. Your rights
Under the GDPR you have the right to:
- access your personal data (Art. 15);
- have inaccurate data corrected (Art. 16);
- have your data erased (Art. 17);
- restrict processing (Art. 18), and be told about corrections, erasure or restriction shared with recipients (Art. 19);
- receive your data in a portable format (Art. 20);
- object to processing based on legitimate interests (Art. 21);
- not be subject to a decision based solely on automated processing that significantly affects you (Art. 22). Paperharp makes no such decisions.
Where processing is based on consent, you can withdraw it at any time; this does not affect processing that happened before.
To exercise your rights, write to info@paperharp.com. Because there is no account and our server does not keep your text, we usually hold little data we can link to you: at most the monthly character counters stored under a hash of your user ID. Most of your data is on your device, and you can view and delete it yourself in the app.
You also have the right to lodge a complaint with a data protection supervisory authority, in particular in the EU country where you live, work or where the alleged infringement took place. A list of EU authorities is available from the European Data Protection Board. In Türkiye, the competent authority is the Personal Data Protection Authority (KVKK).
12. Children
Paperharp is not directed to children under 13, or under the minimum age set by law in their country in the EU. We do not knowingly collect personal data from children. If you believe a child has sent us personal data, for example by email, please contact us and we will delete it.
13. Security
We keep the amount of data that leaves your device small. Your books are stored inside the app’s storage on your device. Natural voice requests are sent to our server over an encrypted connection (HTTPS), our server does not keep the text, and rate limits protect it from abuse. No method of transmission or storage is completely secure, so we cannot guarantee absolute security. Keeping your device’s software up to date and protecting it with a passcode also helps protect your books.
14. Changes to this policy
We may update this policy, for example when we add features such as the natural voice or subscriptions. We will change the “Last updated” date above, and for significant changes we may also let you know in the app or on this website.
15. Contact
Muhammed Tunahan Yollar (Paperharp)
Email: info@paperharp.com