Sandbook

Private · nothing uploaded

A text-to-speech app that never uploads your documents

Sandbook is a free, private text-to-speech app for iPhone and iPad: your documents are never uploaded, because the voice model runs on the phone. Your books, scans, text and voices stay on your device, and there is no account and no ads.

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 that does not upload my documents to the cloud?

Sandbook reads EPUB, PDF, web pages, scanned pages and pasted text aloud with the Kokoro-82M speech model running on the iPhone itself, so the document never leaves the device. Books are copied into the app's own sandbox and read there, in 41 voices across six languages. There is no account and no ads, and what you read never leaves your device. It is free, and needs iOS 18.2 on an iPhone XS, iPhone XR or later, or an iPad with an A12 chip or later.

Last verified November 7, 2026

What “private” means on this page

Most reading apps describe themselves as private and mean a promise: a policy page saying your text is not retained, or is deleted after processing, or is not used for training. Sandbook's version is narrower and checkable. The voice is generated on your iPhone, so the text being read never has to go anywhere. Opening a book, extracting its text, recognizing a scanned page, generating speech and exporting audio all happen on the device. It is not a retention policy; it is the absence of the upload a retention policy exists to govern.

There are two ways to check that without taking our word:

  • Turn on airplane mode and read a book. Speech generated on a server cannot survive that test. The only thing that waits for a connection is adding a new web page.
  • Scan a page with the radios off. The text is recognized and read aloud anyway, because recognition runs on the device too.

The formal version lives in the privacy policy.

What stays on your iPhone

Everything you read, is the short answer. The parts, by name:

  • The model and the voices. Kokoro-82M, open-weight neural TTS with 82 million parameters, is bundled inside the app download along with the 41 voices. Nothing unpacks from a server on first launch, and there is no activation step and no API key.
  • Your files. A book you import, whether EPUB, PDF, Word, TXT, RTF, Markdown or HTML, is copied into Sandbook's own sandbox directory. Deleting it from the library removes that copy and does not touch your original.
  • Your scans. Pages you scan with the camera, photos or screenshots you pick or share from another app, and the pages of a scanned PDF are recognized on the device with Apple's own text recognition. The camera is used only while you scan.
  • Your text. Extracted chapters, PDF text layers, typed text, pasted text and clipboard text are processed in memory and spoken from there, synthesized on the device at 24 kHz, sentence by sentence. The same model also runs in a browser tab on a computer, if you want to test the voices on a private sentence first; that text stays in the tab as well.
  • Your reading position and voices. Where you stopped in each book, the voice and speed each book remembers, your favorite voices and your reading options sit in local app preferences. There is no iCloud sync and no cross-device handoff: a genuine limitation, and the reason none of it is anywhere else.
  • Your exports. M4A audio you export goes wherever you send it through the iOS share sheet, and nowhere else.
  • Web pages you add. With Paste a link or a share from Safari, Sandbook fetches that page, and its cover image or icon, directly from the site that hosts it, like a browser would. No Sandbook server receives the address or the article, and once it is in your library it reads offline like any other book.

What running everything locally costs you

  • Six voice languages, not dozens. 41 voices, 28 of them US and UK English, plus Spanish, French, Italian, Brazilian Portuguese and Hindi. A hosted reader can offer far more.
  • iOS 18.2 and an iPhone XS, iPhone XR or later, or an iPad with an A12 chip or later. A hosted reader runs on older hardware, because it is not running a neural model on your phone.
  • A large download, and no iCloud sync, no account recovery and no library that survives the device.

If any of those matter more than the upload question, the alternatives page compares the readers honestly, including other on-device readers that do several things Sandbook cannot. What running a model on the phone means in general is on the local TTS page.

How cloud text-to-speech apps differ

One generic fact separates the two architectures: if the speech is synthesized on a server, then the text reached that server. What happens to it afterwards is a policy question that every vendor answers for itself; whether it was transmitted at all follows from where the model runs. Each of these is the vendor's own description of where its speech is generated, read on September 14, 2026:

  • Speechify. Mostly on Speechify's servers. On 18 December 2025 Speechify added on-device iOS voices that run locally through Apple's Core ML.
  • NaturalReader. The AI voices are generated on NaturalReader's servers; the Free Voices tier uses the device's own pre-installed system voices.
  • ElevenReader. On ElevenLabs' servers. Audio is downloaded for offline use rather than synthesized on the device.

That list makes no claim about any competitor's document handling or retention; for that, read each vendor's own privacy policy.

Questions about what leaves the phone

Does Sandbook upload my PDFs?

No. A PDF you open is copied into Sandbook's own sandbox directory on the device, and the speech is generated there, on the iPhone itself, sentence by sentence as you listen. There is no upload step, no server-side conversion and no queue. The same is true of EPUB, Word (.docx), TXT, RTF, Markdown and HTML files, of pages you scan with the camera, and of anything you type or paste.

Does it need an account?

No. There is no sign-up screen, no sign-in screen, no email field and no Apple sign-in anywhere in Sandbook. There is also nothing to create an account against: no library on anyone's infrastructure, no sync service, no subscription state. Your reading position, each book's voice and speed, and your favorite voices are stored in local app preferences on the phone, and deleting the app deletes all of it.

Does it work in airplane mode?

Yes, and turning airplane mode on is the fastest way to check this page yourself. The model and all 41 voices ship inside the app download, so EPUB, PDF, Word (.docx), TXT, RTF, Markdown and HTML files, scanned pages, typed or pasted text and clipboard text all read normally with the radios off. The one thing that needs a connection is adding a new web page, because the page has to be fetched before it can be spoken.

What happens when I add a web page?

Sandbook requests the page you chose, and its cover image or icon, directly from the site that hosts it, the same request your browser would make if you opened the link. The app keeps the article and drops the menus, ads and cookie banners on the device, saves it to your library, and reads it locally from then on, offline too. The article text is never forwarded to a Sandbook server, because there is no Sandbook server in that path.

Sources

Every statement about another app on this page is that vendor's own description of where its speech is generated.

  1. Speechify — “Speechify Launches On-Device iOS AI Voices” — speechify.com · checked September 14, 2026
  2. NaturalReader Help Centre — Voices, Languages, and TTS Limits — naturalreaders.com · checked September 14, 2026
  3. ElevenLabs Help Centre — How do offline downloads work? — elevenlabs.io · checked September 14, 2026

Check it in airplane mode

Install it, turn the radios off, and read a whole chapter. A reader that generates speech on a server cannot do that; one that carries the model can. Free, no account, and 41 voices made on your iPhone.

iOS 18.2 or later · iPhone XS / XR or later · iPad with A12 or later · Mac with M1 or later, macOS 15.2

Or try the voices in your browser first