Support

Visit the MurMur page on the Chrome Web Store, click "Add to Chrome", and confirm the installation. The MurMur icon will appear in your browser toolbar. No account or sign-up is needed.

Navigate to any webpage or Google Doc, then click the MurMur icon in your toolbar or press Alt+Shift+R. MurMur will extract the article content and begin reading it aloud with word-by-word highlighting.

Yes. When you open a Google Doc, a "Read Aloud" button appears in the toolbar menu bar. Click it to start reading. You can also use the extension icon or keyboard shortcut as usual.

Open MurMur on any page, then use the Language and Voice dropdowns in the control panel. Select a language first to filter available voices, then choose your preferred voice. Your selection is saved automatically across sessions.

Chrome voices are provided by your browser and may require an internet connection for some high-quality options.

Piper AI voices are neural network voices that run entirely in your browser using WebAssembly. They work offline after the initial model download and produce natural-sounding speech without sending any data to external servers.

MurMur supports 30+ languages including English, Spanish, French, German, Portuguese, Italian, Dutch, Chinese, Japanese, Korean, Hindi, Arabic, Russian, and many more. Piper AI voices add coverage for less common languages like Bulgarian, Catalan, Czech, Danish, Finnish, Georgian, Greek, Hungarian, Icelandic, Kazakh, Nepali, Norwegian, Romanian, Serbian, Slovak, Slovenian, Swahili, Swedish, Ukrainian, and Vietnamese.

Use the speed slider in the control panel (0.5x to 3x) or press ↑ / ↓ arrow keys to adjust in 0.25x increments. For Piper AI voices, speed changes are applied at the synthesis level so pitch stays natural at any speed.

Overlay mode opens a clean, distraction-free reader view on top of the page with comfortable typography and a dark background. Ideal for focused listening.

Inline mode highlights words directly on the original page without changing the layout. Great for following along with the original formatting, images, and context. Switch between modes from the control panel.

Shortcut	Action
Alt+Shift+R	Toggle MurMur on/off
Space	Play / Pause
←	Previous paragraph
→	Next paragraph
↑	Increase speed
↓	Decrease speed
Esc	Close MurMur

Some pages use complex JavaScript rendering that makes text extraction difficult. MurMur also cannot run on Chrome's built-in pages (chrome://, chrome-extension://) or the Chrome Web Store due to browser security restrictions. If MurMur can't detect a structured article, it falls back to reading visible text blocks from the page body.

Make sure your system volume is turned up and no other audio is blocking output. If using Chrome voices, try selecting a different voice from the dropdown. If using Piper AI voices, the first use requires downloading the voice model which may take a few seconds. Check that your browser is not muted in your operating system's sound settings.

Word-level highlighting depends on timing events from the TTS engine. If highlighting drifts, try reducing the reading speed or switching to a different voice. Chrome's built-in voices generally provide the most accurate word-timing events.

The "Read Aloud" button is injected after the Google Docs toolbar renders. Try refreshing the page. If the button still doesn't appear, make sure MurMur is enabled in chrome://extensions and that you're on a docs.google.com/document/ URL.

No. All text processing happens locally in your browser. MurMur does not track, store, or transmit any of your browsing or reading data. The only network request MurMur makes is to download Piper voice model files from HuggingFace when you first use a Piper voice.

Chrome's built-in voices may require an internet connection depending on the voice. Piper AI voices work fully offline after the initial voice model download. Your preferences (speed, voice, pronunciation rules) are stored locally and always available offline.

activeTab — Access the current page to extract article text when you click the icon.
scripting — Inject reader scripts for text extraction and highlighting.
storage — Save your preferences (speed, voice, mode, pronunciation rules).
tts — Access Chrome's built-in text-to-speech engine.
offscreen — Run Piper TTS WebAssembly in a background document.
huggingface.co — Download Piper voice models from HuggingFace's open model repository.