← Back to Varterm

Privacy Policy

Effective date: May 8, 2026

Overview

Varterm provides text-to-speech features across our web app and browser/editor extensions. This policy explains what data is processed, why it is used, and what choices you have.

What We Process

  • Text you choose to read aloud (selected text, pasted text, or page content).
  • Basic extension/app settings (voice, speed, and markdown preference).
  • Basic analytics data about visits and feature usage trends.
  • Optional feedback details you submit (message, optional email).
  • Operational request data needed to deliver audio responses.

Offline Voices

The web reader offers an offline voice option. When you select it, speech is synthesized by Piper inside your browser and the text you paste never leaves your device. The first time you use a given offline voice, the browser downloads that voice model (about 63MB) from Hugging Face and caches it locally; the download request reveals your IP address and which voice you chose to Hugging Face, but not your text. The speech runtime itself is served from varterm.com. You can delete cached voice models at any time with the "Remove downloaded" control next to the voice list.

Third Parties

  • Microsoft: cloud voices are synthesized by the Microsoft Edge neural speech service, so the text you submit is sent there to produce audio.
  • Google Analytics and Google Fonts: page-visit measurement and web fonts. These receive request metadata such as IP address and user agent. They do not receive the text you submit.
  • Vercel: hosting, so all requests to this site transit their infrastructure.
  • Hugging Face: only when you download an offline voice model.
  • Formspree: only if you submit the feedback or support form.
  • Stripe: only if you choose to donate; payment happens on their site.
  • ElevenLabs and Anthropic: used by the premium API and the extension document question feature, not by the web reader.

How Data Is Used

  • To generate audio output for text-to-speech playback.
  • To apply your saved settings consistently across sessions.
  • To understand aggregate product usage and improve reliability.
  • To operate and troubleshoot service reliability.

We do not sell personal data, do not use your content for unrelated advertising purposes, and do not use the text you submit to train models.

Chrome Extension Permissions

  • activeTab: Used after user action to access the current tab text for read-aloud.
  • contextMenus: Adds right-click actions for reading selection or page content.
  • scripting: Injects local extension scripts to read selected/page text and show controls.
  • storage: Saves preferences like voice, speed, and markdown mode.
  • Host permissions: Allows requests to Varterm API endpoints for speech generation.

Remote Code Statement

The Chrome extension does not execute remotely hosted code. Network requests are used only to send text for speech generation and receive audio output.

Data Retention

We retain data only as long as needed to provide the service, secure operations, and comply with legal obligations.

Your Choices

  • You can disable or uninstall the extension at any time.
  • You can clear browser extension data and stored settings in Chrome.
  • You can choose browser voice mode for local playback when available.

Contact

For privacy questions, use the support form.