Paste a link
Copy a public article URL, paste it into PageVoice and choose Prepare article.
PageVoice help centre
Good words.
Better heard.
Find answers about preparing web articles, choosing voices, moving between chapters and listening when you are away from a connection.
Quick start
Listen in three steps
PageVoice works with readable article pages available through an ordinary HTTP or HTTPS link.
Choose your sound
Select Apple speech, an installed voice and a listening pace. Piper is an optional advanced server preview.
Press Play
Follow the text, move by paragraph and use background or lock-screen playback controls.
Original preserves the extracted wording. AI rewrite deliberately prepares different wording for listening and can make mistakes. Compare important details with the source.
Troubleshooting
Common questions
Open a question below for the shortest route back to listening.
PageVoice cannot find a readable article
Check that the link opens and that your device has a connection. PageVoice reads public HTML article pages; it does not bypass paywalls, sign-ins or website restrictions.
Choose Browse website in app, navigate using the website’s own controls, then tap Read this page. Some highly interactive pages, infinite feeds, PDFs and video-only pages may not provide readable article text.
The next page or chapter is not detected
PageVoice intentionally follows only a confident, same-website next-page link. Ambiguous links, recommendations, backwards links and JavaScript-only buttons are rejected to avoid opening the wrong page.
Use Browse website in app, tap the site’s own Next control, then choose Read this page. Auto-next is off again each time you launch the app.
I want a more natural Apple voice
On your device, open Settings → Accessibility → Read & Speak → Voices and download an enhanced or premium voice where available. Return to PageVoice, reopen the voice picker and use Preview before selecting it.
Available voices vary by language, region and operating-system version. Siri’s private voice models are not available to PageVoice.
Apple Intelligence features are unavailable or an AI rewrite failed
Summaries and AI rewrite require a supported device, language and region with Apple Intelligence enabled and ready. Original reading and Apple speech still work without those features.
A rewrite is rejected if PageVoice detects missing content, major length changes, changed numbers or changed quotations. Retry it, or choose Original to hear the extracted article without an AI rewrite.
Piper will not connect or starts slowly
Piper is optional and requires access to a compatible authenticated HTTPS server. In Piper server settings, enter the server origin, import or save its access token, then use Check connection. After restarting your device, unlock it once before locked-screen Piper playback.
A server model change or weak connection can delay the first passage. Piper tracks paragraphs, not individual words. Changing voice cancels the old generation and restarts the current paragraph; changing it while paused stays paused.
An offline chapter is incomplete
Offline chapter preparation is available when Piper is configured. Keep PageVoice open until every chapter you need says Ready offline. Wi-Fi is the default; preparation can be allowed over mobile data, while Low Data Mode pauses it.
Changing server, voice or reading style creates different audio. Incomplete chapters can play the audio already prepared, but missing passages still need a connection. PageVoice never silently deletes ready downloads to make space; preparation stops at its storage limit.
Playback stopped while the screen was locked
Apple speech and already prepared audio can continue in the background. Piper may pause if it runs out of prepared audio, the network fails or iOS ends background work. Open PageVoice, review the message and press Play to retry.
Calls, Siri, route changes and headphone disconnection can interrupt playback. PageVoice does not resume unexpectedly after an interruption; press Play when you are ready.
How do I remove saved pages, downloads or a Piper token?
- Remove one saved page from its website folder, delete a complete folder, or use Clear listening shelf.
- Open Offline listening to clear downloaded Piper chapters separately.
- Use Forget token in Piper server settings to remove that server origin’s token from this device.
- If you imported a plaintext token file, delete that file from Files and Recently Deleted after it has been saved to Keychain.
Useful to know
Reading, voices and travel
Apple speech
Runs on your device, supports installed iOS voices, highlights the current word and offers playback from 0.75× to 1.5×.
Optional Piper preview
Uses the voice catalog supplied by your configured authenticated server. Narration is remote, with paragraph tracking and no accurate word highlighting.
Long reads
Move manually or use Auto-next where a reliable next-page or next-chapter link is available. The in-app website is the manual fallback.
Your listening shelf
Saved pages are grouped by website and remember your paragraph. Delete one page, a whole website folder or the complete shelf whenever you choose.
For advanced users
Run your own Piper server
Piper is optional. It is for people who control a compatible speech server and want PageVoice to generate audio through it. It is not required for Apple speech, article reading or Apple Intelligence features.
1. Install the speech engine
Start with the upstream Piper installation guide, then use its HTTP-server guide for a local synthesis test.
2. Choose voices carefully
Use the official Piper voices guide, listen to the voice samples, then find model cards in the voice catalogue. Review each model’s licence before use or redistribution.
3. Add a secure PageVoice adapter
PageVoice needs a compatible HTTPS service in front of Piper. It uses authenticated versioned API routes, bounded text requests, cancellation and WAV audio; the upstream Piper web service alone is not a drop-in PageVoice endpoint.
4. Configure PageVoice
In Piper server settings, enter only your HTTPS server origin, save or import its access token, then use Check connection. Never share a token in a URL, screenshot, email or support request.
Self-hosting is a security and privacy responsibility. The stock Piper HTTP server is useful for local testing, but is not a drop-in PageVoice endpoint and should never be exposed publicly without TLS and authentication. Use a compatible HTTPS bridge with bearer-token authentication, request limits and a short bounded queue. Decide what, if anything, your server logs or retains and reflect that accurately in your privacy policy. Piper does not currently give PageVoice accurate source-word timings, so the app keeps paragraph tracking for Piper and word highlighting for Apple speech.
Privacy at a glance
Where your reading goes
These points explain normal app behaviour. They are not a replacement for the separate privacy policy supplied with the public App Store listing.
On your device
Apple speech and Apple Intelligence processing run on your device. Shelf text, bookmarks and settings are stored locally. PageVoice does not require an app account.
Websites you open
Preparing a link contacts that website. PageVoice uses a temporary in-app web session rather than sharing your Safari sign-ins, and it does not bypass a site’s access controls.
When you choose Piper
Readable narration text is sent over authenticated HTTPS to the server you configured. PageVoice does not send that server the website’s HTML, and the token is kept in the iPhone Keychain.
This support page
This page contains no analytics, advertising, cookies, external fonts or third-party scripts.
Still need help?
Contact PageVoice support
Include your device model, iOS or iPadOS version, PageVoice version and a short description of what happened. Never send a Piper token or private article text.
Legal notices
Copyright and acknowledgements
PageVoice
© 2026 Tony Ashton. All rights reserved.
Mozilla Readability
PageVoice includes unmodified Mozilla Readability 0.6.0. Copyright © 2010 Arc90 Inc. Licensed under the Apache License 2.0. See the upstream source and the licence included with the app.
Piper
Piper is a third-party open-source project, licensed under GPL-3.0, and is not affiliated with or endorsed by PageVoice. Self-hosting operators must review the separate licence for each voice model.
Apple trademarks
Apple, iPhone, iPad, iPadOS and Apple Intelligence are trademarks of Apple Inc., registered in the U.S. and other countries and regions. PageVoice is not affiliated with Apple Inc.