Privacy

Scope

This page describes the current app code. A production hosting provider and your browser or platform may have separate data practices. Privacy questions can be sent to info@drop2hear.com.

Pasted text and selected files

The app reads pasted text and selected PDF, image, DOCX, TXT and Markdown files in the current tab. There is no app document-upload endpoint or account system. The app does not intentionally send these files or extracted text to a server.

PDF and OCR processing

PDF extraction and OCR run in your browser. PDF.js, Tesseract worker files, and the selected English, Spanish, French, German, Hindi, or Portuguese OCR data are served as same-origin site assets when needed. OCR recognition may be inaccurate.

Instant Listen voices

Instant Listen calls the browser Web Speech API. Available voices and their handling are controlled by your browser or operating system. Some voices may use online services, which could transmit spoken text to the provider. Choose a voice known to work offline if local speech matters to you.

Natural Audio

Natural Audio runs speech generation in a browser worker. The app sends model-asset requests to Hugging Face on first use; it does not send your document text to that service for generation. These requests can reveal ordinary network metadata such as your IP address to the asset host.

Model caching

The Transformers.js library may store downloaded model assets in your browser Cache Storage under a transformers cache. Those cached assets are public model files, not your document text. You can remove them through your browser site-data controls.

App-shell caching

The Angular service worker may cache public app-shell and feature assets, including an OCR language pack after it is requested, for offline reuse. Its configuration does not cache your uploaded documents or generated audio. Offline Natural Audio is not guaranteed even after an initial download.

Current-page memory

Extracted text and generated audio are held in the tab while you work. Generated audio is released when you clear or replace content; a WAV you download remains on your device. The current extracted text can also be kept in local browser storage for the resume feature described below.

Browser storage

The resume feature stores one current reading in this browser using IndexedDB, including extracted text, source name and size, approximate position, speed, language and voice preferences, and bookmarks. The selected OCR language is stored separately as a small local preference. This data is not synchronized or sent to a Drop2Hear backend. Loading different content replaces it. Clear saved reading or your browser’s site data to remove it. The app does not set application cookies or include analytics or advertising trackers.

Cookies

The current app code does not set application cookies. Browser, platform, or future hosting services may have their own practices; this page must be updated if deployment changes that.

Analytics

The current app does not include analytics or advertising trackers. This would need a new review and updated notice if added later.

Static hosting requests

A static host receives requests for pages and assets and may retain ordinary access logs such as IP address, time, and requested URL. The hosting provider and its retention settings must be reviewed when deployment is chosen.

No accounts or payments

The current app has no user accounts, payment flow, database, or server-side document history.

Your controls

You can stop playback or generation, use Clear saved reading to remove the current reading and bookmarks from IndexedDB, remove all cached site data in your browser, and delete WAV files you downloaded.

Security limits

Browser-first processing reduces app-side document transfers, but no website can guarantee absolute privacy or security. Browser extensions, platform voices, and the device itself remain outside this app’s control.

Children and sensitive material

The tool is a general-purpose beta and is not designed as a child-specific service. Consider your browser voice choice and device security before using sensitive material.

Changes and contact

This page should be reviewed and updated when hosting, analytics, accounts, or processing behavior changes. Privacy questions can be sent to info@drop2hear.com.