Neural Voices by Jonas

Neural Voices reads web pages and PDFs aloud with natural-sounding neural voices (Piper). The voices run inside your browser, so there is no server, no account and no cloud: the text you listen to never leaves your computer.

WHAT IT DOES

• Reads the part you want: the whole article without menus, sidebars and footers, a selection, or from any sentence you click.
• Follow along: the sentence is shaded and the word being spoken is marked. The page scrolls with the reading.
• Any language: each paragraph's language is recognised from its words, so a page that mixes languages changes voice as it goes. 170 voices in 49 languages; press ▶ Try in Settings to hear one before adding it.
• PDFs: open a PDF from the web or your computer; it's shown in the add-on's own viewer and read like a page.
• Natural speed: 0.75× to 2×, changed inside the voice, so it sounds like a person talking faster, not a sped-up recording.
• Your pronunciations: if a voice says a name wrong, select it, right-click "Fix pronunciation" and write how it should sound.
• Works offline: a voice downloads once (20–140 MB, most about 60 MB) and is kept in the browser.

CONTROLS

A small player on the page: play/pause, previous/next sentence, speed, volume, voice. Shortcuts: Alt+Shift+R read / pause, Alt+Shift+X stop, Escape stops.

PRIVACY

No tracking, no analytics. The only network use is downloading the voice files you choose from Hugging Face (the Piper voice collection) and, when you press ▶ Try, playing samples from the Piper project's website.

GOOD TO KNOW

The first sound takes a few seconds while the voice loads; after that it reads ahead. A reasonably modern computer is needed (one CPU core free). Voices are made by the Piper community; each voice's licence is shown in Settings.

Open source (GPL-3.0): https://github.com/JCA374/read-aloud