Offlinetts

OfflineTTS reads web pages, PDFs, and selected text aloud using high-quality AI voices — completely offline and private. No data ever leaves your browser.

Features

- 4 Natural AI Voices — Choose from Male 1, Male 2, Female 1, or Female 2
- Read Full Pages — Extract and read article text with one click
- Read Selected Text — Highlight text, right-click, and select "Read with OfflineTTS"
- PDF Support — Read PDF documents directly in your browser
- Speed Control — Adjust playback from 0.5x to 2.0x
- Pause & Resume — Stop anywhere and continue later
- Keyboard Shortcuts — Alt+P (play/pause), Alt+O (stop), Alt+. (skip forward), Alt+, (skip back)

Privacy First

Everything runs locally on your device using ONNX Runtime. Your browsing data and text content are never sent to any server.

Lightweight

Runs entirely on CPU — no GPU required.

Open Source

OfflineTTS is free and open source under the Apache 2.0 license. View the source code and contribute at github.com/hkdb/offline-tts

Credits

Powered by Supertonic (github.com/supertone-inc/supertonic) for high-quality neural TTS voices.

Early Development

OfflineTTS is in active development (v0.1.0). Currently supports English (tested) and Korean (not-tested). Feedback and contributions welcome!