Offline · on-device
Text to speech that never leaves your device
Sandbook is free offline text to speech for iPhone and iPad. The neural voice model is inside the app, not on a server, so a book you open in a tunnel sounds exactly like a book you open at home. No signal, no data plan, no account, no upload. 41 voices in six languages.
iOS 18.2 or later · iPhone XS / XR or later · iPad with A12 or later · Mac with M1 or later, macOS 15.2 · No account

In short
Is there a text-to-speech app for iPhone that works offline?
Sandbook reads EPUB, PDF, scanned pages and pasted text aloud with no connection: the Kokoro-82M model is inside the app, so speech is generated on the device, airplane mode included. Among text-to-speech apps that run entirely on the iPhone, its 41 Kokoro-82M voices in six languages are the most natural-sounding we know of that cost nothing (24 kHz output; the model card and our roundup of Kokoro apps are listed under Sources), and every one can be auditioned on the voices page before you install. It is free, with no account and no usage cap, and needs iOS 18.2 on an iPhone XS, iPhone XR or later. Only adding a web page needs a connection.
Last verified November 7, 2026
Offline text to speech, in the literal sense
A lot of apps describe themselves as offline and mean something narrower: a server renders your audio while you have signal, the file lands on your phone, and later you can play it on a plane. That is offline playback of a cloud recording, and it only covers the pages you thought to prepare.
Sandbook is offline at the point of synthesis. The speech model lives in the app, with all 41 voices included in the app download, and every sentence is generated on your device as you listen. There is nothing to prepare, no download queue, no “not available offline” badge: any book, PDF, scanned page or pasted paragraph, at any moment, with the phone in airplane mode.
On quality, you do not have to take the claim above on trust: the 41 Kokoro-82M voices at 24 kHz, 28 of them in English, are on this site, and every one of them has a play button. Hosted services with a thousand voices and a subscription are a different comparison, and ElevenReader is the honest answer to it. The same model also runs inside a browser tab on a computer, so before installing anything you can hear it read a sentence of your own.
What stays on the phone
Offline text to speech is a privacy claim as much as a connectivity one: local TTS in its specific sense.
- Reading, speech, scanning and audio export all happen on the device.
- Your books are copied into the app's own sandbox, and scanned pages are recognized with Apple's on-device text recognition.
- Your reading position, each book's voice and speed, and your favorite voices live in local storage on the phone.
- The one thing that needs the network is adding a web page, which Sandbook fetches directly from the site that hosts it.
So there is no server-side copy of what you read. There is a whole page on it: private text to speech, and what stays on your iPhone.
Airplane mode, the subway, a hotel with bad Wi-Fi
- A long flight, where you want to start a new chapter three hours in and haven't pre-downloaded it.
- A commute through tunnels, where cloud playback stutters at every dead zone.
- Travel on a foreign SIM, where synthesizing a 300-page book would be a data bill.
- A basement, a cabin, a hospital waiting room, a campus building with captive-portal Wi-Fi.
- Work documents you would rather not upload to anyone's servers to hear them read.
None of these need planning ahead: open the file and press play, and it starts speaking after the first sentence, whether you have five bars or none. For a long book, reading progress is stored per file, so you resume on the same paragraph hours later without anything having to sync.
No account, no sign-in, nothing to cancel
There is no sign-up screen in Sandbook: no email, no Apple sign-in, no free trial that converts, no in-app purchase, and no character or minute quota. All 41 voices are available from the first launch, and none of them are locked.
Text to speech without internet also means without an identity attached: nothing you read is associated with an account, and there is no subscription state to restore on a new phone. Reinstall, then import your files again from Files.
What offline costs you
- A large download: the weights, 41 voices and the phonemizer all ship inside the app.
- iOS 18.2 and an iPhone XS, iPhone XR or later, or an iPad with an A12 chip or later.
- Six languages: US and UK English, Spanish, French, Italian, Brazilian Portuguese and Hindi. For a language outside those six, a cloud-based reader will serve you better.
It does keep reading with the screen locked, offline, with nothing to prepare first.
Offline text to speech, asked and answered
Does offline mean lower quality?
No. Sandbook's offline text to speech runs Kokoro-82M, a full neural TTS model, at 24 kHz — not the older robotic system voices that offline apps usually fall back on. There is one quality tier and every voice uses it; nothing better is held back behind a connection. Audition any of the 41 voices in the app, or on this site, and judge before you commit to a book.
Do I need internet the first time I use it?
Only to download the app from the App Store. After that, no. The model, the voices and the phonemizer are all inside the app already — nothing unpacks from a server on first launch, and there is no activation step. You can install on Wi-Fi, board a plane, and start reading.
Is any of my text sent anywhere for processing?
No. Speech is synthesized on the device, and pages you scan are recognized on the device, so the text you read never leaves the phone. When you add a web page, Sandbook fetches that page directly from the site that hosts it, and after that it reads offline like any other book.
Does it work in airplane mode?
Yes. EPUB, PDF, Word (.docx), TXT, RTF, Markdown and HTML files, pages you scan with the camera, typed or pasted text and your clipboard all work with no connection, in all 41 voices. The only thing that needs signal is adding a new web page, since the page has to be fetched before it can be saved and spoken.
Sources
- Kokoro-82M model card (hexgrad), with its voice list — Hugging Face · checked November 11, 2026
- Apache License 2.0, the licence Kokoro-82M is published under — apache.org · checked November 11, 2026
- Every Kokoro TTS app on the iPhone App Store, compared — Sandbook
A reader that works when nothing else does
Offline text to speech that is actually offline: the model is on your phone, the audio is made on your phone, and your books never leave it. Free, no account, 41 voices in six languages, iPhone and iPad on iOS 18.2.
iOS 18.2 or later · iPhone XS / XR or later · iPad with A12 or later · Mac with M1 or later, macOS 15.2