Changelog

1.1.0 — 22 October 2026

The first public release.

• Word-by-word highlighting while the page is read aloud, painted directly in the page.
• Gentle auto-scroll with a visible Scroll along switch. It backs off when you scroll yourself.
• Start anywhere: ⌥-click a paragraph, select a passage and choose Read selection, or right-click and pick Read from here.
• Any language your computer can speak. Auto chooses a voice that matches the page; Japanese, Hindi, and German detection are covered by the test suite.
• Docs-friendly extraction that prefers the main article and finds inner scroll panes.
• Light and dark themes, adjustable speed, voice selection, and a progress counter.
• Keyboard shortcuts for play/pause, stop, next sentence, and previous sentence.
• Local by design: no account, no analytics, no network requests, and no remote code.
• Fallback timing for voices that do not emit word boundaries, and a retry when the browser blocks the first speech attempt.

1.0.0 — internal

Development build used for testing. Not published to the store.

Planned — not shipped

These are ideas, not promises, and they do not have dates:

• Optional cloud voices for more natural narration, opt-in and disclosed before any text leaves the device.
• PDF and EPUB support.
• Per-paragraph navigation, bookmarks, and resume-where-you-left-off.
• A focus mode that dims everything except the current paragraph.

If a future version adds cloud voices, the privacy policy will be updated first and the feature will be off by default.