Read-Aloud Study Reader

Privacy Policy

Effective 5 September 2026 · Publisher: Daniel Clayton

Overview

Read-Aloud Study Reader is a document-listening and study reader. It has no advertising, developer account system or developer-operated analytics. The publisher does not sell personal data.

Information you choose to use

The app can process documents and pasted text, provider credentials, pronunciation entries, narration and visual settings, and export destinations that you select. Documents are opened only after you choose them through Android’s system document picker.

Local processing and storage

Extracted text, rendered PDF page images, document metadata, reading positions, preferences, pronunciation entries, local structural summaries, speech cache and the optional Kokoro model are stored in app-private storage. Android cloud backup and device-to-device transfer are disabled. Provider keys are encrypted using AES-GCM with a non-exportable Android Keystore key.

Local data remains until you remove the relevant item, model, key or cache, clear app storage, or uninstall. Exported files remain at the location you chose and must be deleted there separately. The app has no remote account to delete.

When the app uses the network

  1. Kokoro resources. After you press Download, the app retrieves pinned model, voice, configuration and English dictionary files over HTTPS, verifies their sizes and SHA-256 checksums, and stores them privately. Your document text is not sent with this download; narration then runs on-device.
  2. ML Kit OCR. Google Play services may download or update the on-device OCR model. OCR images and recognised text remain on-device. Google states that ML Kit may send device, app, SDK, performance, event, error and related diagnostic information.
  3. Google Cloud Text-to-Speech. If you explicitly select it, bounded narration passages and your credential are sent over HTTPS to Google Cloud.
  4. Microsoft Azure neural speech. If you explicitly select it, bounded narration passages and your credential are sent over HTTPS to the Azure region you configured.
  5. Installed Android speech. An installed speech engine receives narrated text when you select it. Its supplier controls whether processing is local or networked.

Google, Microsoft and installed speech providers process data under their own terms and privacy notices. Retention, processing location, billing, output rights and acceptable-use rules can vary by provider, account, tier and country. The app publisher does not receive a separate copy of documents, generated speech, credentials or exports from these features.

Purpose and legal basis

Processing is limited to providing the feature you request: document access, extraction, OCR, narration, caching, reading-position restoration, local structural summarisation or export. Where UK or EU data-protection law applies, user-directed processing needed to provide these functions is based on performance of the service contract. Third-party providers describe their own legal bases in their privacy notices.

Sharing, security and control

The publisher does not share data for advertising. Data reaches a cloud speech provider only when you select that provider and use it, except for ML Kit’s documented operational information and model delivery through Google Play services. The app uses app-private storage, HTTPS-only endpoints, Android Keystore encryption, bounded requests and no broad storage permission. No system can be guaranteed completely secure.

Accuracy and user control

Extraction, OCR, pronunciation substitutions, synthetic speech and structural summaries can be incomplete or wrong. Check important information against the original source. You control which documents are imported, whether an optional cloud speech provider receives selected text, and where exported files are stored.

Changes and contact

This policy will be updated when data handling, providers or legal requirements materially change. Privacy and support enquiries: dan6098@hotmail.com.