Lullula – Text to Speech & Read Aloud by Lullula
The read-aloud extension built for listening. Web pages, Google Docs, PDFs and EPUBs, read aloud in natural voices with each sentence highlighted and your place kept. 600+ neural voices in 80+ languages, plus free built-in voices.
Some features may require paymentSome features may require payment
6 Users6 Users
Extension Metadata
Screenshots
About this extension
Lullula is the read-aloud extension built for listening. It reads web pages, Google Docs, text-based and scanned PDFs, and DRM-free EPUB books aloud in natural neural voices, highlights each sentence as it is spoken, and remembers the sentence you stopped on — so you can follow an article with your ears on a commute, while cooking, working out, or resting your eyes.
BUILT FOR LISTENING
Reader views find the article; they were designed for eyes. Your eyes skip a line that isn't for you and back up half a paragraph without thinking. A voice can't, so Lullula puts more around it:
• It starts at the first sentence — not at the top of the page, and not at a note the article leads with.
• The highlight follows the voice — each sentence lights up as it's spoken and the page scrolls to keep it in view. Glance back and you're on the right line.
• Speed, and a step back by sentence — 0.5× to 2× in seven steps; back and forward move one whole sentence, so hearing something again starts where the thought starts.
• It remembers the sentence — close the tab, come back tomorrow, press play. Lullula picks up where you stopped, with a place kept per page.
• Google Docs, PDFs and EPUBs work the same way — same player, same highlight, same saved place on a doc, a paper or a book.
• Scanned PDFs work too — Lullula recognizes the text on each page in your browser, then reads it aloud like any other page.
WHAT YOU GET
• 600+ natural neural voices in 80+ languages — English, Spanish, Chinese, French, Hindi, Arabic, Japanese and more — plus unlimited free built-in browser voices
• Reads a Reddit or Hacker News thread as a conversation, each commenter in their own voice
• AI summaries — get the gist of a long page before you commit to the full listen
• Light and dark themes, and a draggable mini player that stays out of your way
• Listen later — one click saves a page's text to your Lullula library to play later: with Premium the natural-voice audio is prepared ahead of time; on the free plan it plays with your device's own voice
GREAT FOR
• Commuters and multitaskers who'd rather listen to articles than read them
• Students and researchers getting through long PDFs, papers and study guides
• Readers with dyslexia, ADHD, low vision or eye strain — listening keeps the words flowing
• Language learners who want natural pronunciation at an adjustable speed
FREE TO START
Unlimited listening with built-in browser voices, plus 2,000 words of premium neural voice every month once you sign in — AI summaries from the same allowance. Lullula Pro adds a monthly allowance of premium neural-voice listening.
PRIVACY
Built-in browser voices speak locally. A premium voice sends the text you play to Lullula's service to generate the audio. When the reading player appears on a page, a summary is written automatically, so that page's text is sent to Lullula's service and the AI provider that writes it; you can turn summaries off in the extension's settings. If you sign in, your email address and authentication token are used to provide your account and sync your settings, and the pages you play are saved to your reading history (title, address, progress, and a short excerpt around your resume point). If you save a page with Listen later (sign-in required), its title, address and the text Lullula reads aloud are stored with your account so you can play it later; with Premium, that text is also sent ahead of time to Lullula's service and its speech provider to prepare the audio. If you connect your own text-to-speech provider (ElevenLabs, OpenAI or an OpenAI-compatible service, Azure Speech, Google Cloud Text-to-Speech, Amazon Polly or Alibaba Cloud Qwen-TTS) with your own key, the text you play with its voices goes straight from your browser to that provider under its terms, and the key stays in your browser, never sent to Lullula; summaries, if on, still send the page to Lullula as above. Lullula does not sell your data, build advertising profiles, or monitor unrelated browsing. Built by a small team. Full policy, naming every service that receives text: https://lullula.io/privacy
Install Lullula and listen to the web, your documents and your books today.
BUILT FOR LISTENING
Reader views find the article; they were designed for eyes. Your eyes skip a line that isn't for you and back up half a paragraph without thinking. A voice can't, so Lullula puts more around it:
• It starts at the first sentence — not at the top of the page, and not at a note the article leads with.
• The highlight follows the voice — each sentence lights up as it's spoken and the page scrolls to keep it in view. Glance back and you're on the right line.
• Speed, and a step back by sentence — 0.5× to 2× in seven steps; back and forward move one whole sentence, so hearing something again starts where the thought starts.
• It remembers the sentence — close the tab, come back tomorrow, press play. Lullula picks up where you stopped, with a place kept per page.
• Google Docs, PDFs and EPUBs work the same way — same player, same highlight, same saved place on a doc, a paper or a book.
• Scanned PDFs work too — Lullula recognizes the text on each page in your browser, then reads it aloud like any other page.
WHAT YOU GET
• 600+ natural neural voices in 80+ languages — English, Spanish, Chinese, French, Hindi, Arabic, Japanese and more — plus unlimited free built-in browser voices
• Reads a Reddit or Hacker News thread as a conversation, each commenter in their own voice
• AI summaries — get the gist of a long page before you commit to the full listen
• Light and dark themes, and a draggable mini player that stays out of your way
• Listen later — one click saves a page's text to your Lullula library to play later: with Premium the natural-voice audio is prepared ahead of time; on the free plan it plays with your device's own voice
GREAT FOR
• Commuters and multitaskers who'd rather listen to articles than read them
• Students and researchers getting through long PDFs, papers and study guides
• Readers with dyslexia, ADHD, low vision or eye strain — listening keeps the words flowing
• Language learners who want natural pronunciation at an adjustable speed
FREE TO START
Unlimited listening with built-in browser voices, plus 2,000 words of premium neural voice every month once you sign in — AI summaries from the same allowance. Lullula Pro adds a monthly allowance of premium neural-voice listening.
PRIVACY
Built-in browser voices speak locally. A premium voice sends the text you play to Lullula's service to generate the audio. When the reading player appears on a page, a summary is written automatically, so that page's text is sent to Lullula's service and the AI provider that writes it; you can turn summaries off in the extension's settings. If you sign in, your email address and authentication token are used to provide your account and sync your settings, and the pages you play are saved to your reading history (title, address, progress, and a short excerpt around your resume point). If you save a page with Listen later (sign-in required), its title, address and the text Lullula reads aloud are stored with your account so you can play it later; with Premium, that text is also sent ahead of time to Lullula's service and its speech provider to prepare the audio. If you connect your own text-to-speech provider (ElevenLabs, OpenAI or an OpenAI-compatible service, Azure Speech, Google Cloud Text-to-Speech, Amazon Polly or Alibaba Cloud Qwen-TTS) with your own key, the text you play with its voices goes straight from your browser to that provider under its terms, and the key stays in your browser, never sent to Lullula; summaries, if on, still send the page to Lullula as above. Lullula does not sell your data, build advertising profiles, or monitor unrelated browsing. Built by a small team. Full policy, naming every service that receives text: https://lullula.io/privacy
Install Lullula and listen to the web, your documents and your books today.
Rated 0 by 0 reviewers
Permissions and data
Required permissions:
- Access browser tabs
- Access browser activity during navigation
- Access your data for all websites
Optional permissions:
- Access your data for all websites
More information
- Add-on Links
- Version
- 2.3.1
- Size
- 32.12 MB
- Last updated
- 5 hours ago (Nov 14, 2026)
- Related Categories
- License
- All Rights Reserved
- Privacy Policy
- Read the privacy policy for this add-on
- Version History
- Add to collection
• Browsing activity: signed-in playback syncs the selected page/document title, URL, favicon, content type, progress, duration and reading position for history and resume. Listen later stores the chosen item's title and URL. A signed-in saved-state lookup sends the current page URL without storing that lookup with the account. Following a page stores its address and lets our server check its public content or feed on a schedule, without your cookies.
• Website content: signed-in resume sync includes a short excerpt (up to 80 characters). Premium speech, summaries you request, and Listen later send the relevant text to Lullula; premium saved items can prepare audio ahead of time. Page analysis can send a language sample (up to 500 characters), an article outline and web-page formulas. Scanned PDF pages and equations can be sent as images for recognition. When signed in with Describe figures with AI on, PDF figures or their pages, captions, a nearby sentence and reading language can be sent ahead of playback to Lullula and Anthropic for description. When Read image is available, chosen images are redrawn without photo metadata. The image, language and account token or trial ID go to Lullula; Anthropic receives the image and language. Save stores the images and reading in your private library; the full policy explains other retention. Own-key TTS sends speech text directly to the selected provider; the key and those audio requests do not reach Lullula. Analysis, summaries, history and saves still follow their disclosed flows.
• Authentication information: an account token is used for sign-in and account/premium access.
• Technical and interaction data: first-party reports cover installation, playback request/first engine start/voice selection, listening duration/outcome, start latency, fixed stop/refusal codes, trial/entry/welcome actions. They include technical voice/model IDs, language/compatibility tags, script/length buckets and requested playback settings; native labels are locally derived IDs. No raw text/audio, page URL/title, personal labels or provider keys/endpoints. Random IDs may link to verified accounts. Up to 500 pending reports retry for 24 hours. Optional player-design measurement remains disabled.
Piper and Matcha synthesize free speech on-device. First Play, Preview or explicit model load downloads model data from Hugging Face and its CDN for offline caching. The download reveals the public model choice and ordinary connection metadata (IP address and browser details), not reading text, page URL, account token or generated audio. Signed-in history and other selected features still work as described above.
Lullula does not monitor unrelated browsing. Store builds do not capture screenshots of visited web pages; PDF and chosen-image flows are described above.