Local TTS · on-device
Local TTS, running on your iPhone
Sandbook is a free local TTS app for iPhone, iPad and Mac: the Kokoro text-to-speech model runs on the device, so your text is turned into speech without being sent to a server. 41 voices in six languages, airplane mode included, with no account and no meter.
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
What is a good local TTS app for iPhone?
Sandbook runs the Kokoro-82M text-to-speech model on the iPhone itself, so books, PDFs, web pages, camera scans and pasted text are read aloud without any text being sent to a server. It has 41 voices in six languages, highlights each word as it is spoken, and works in airplane mode. It is free, with no in-app purchase, no account and no character limit, 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 “local TTS” means
TTS is text to speech: software that turns written words into a spoken voice. Local TTS means the model that does the turning runs on your own device. Cloud TTS means it runs on someone else's computer: your text is sent over the network, a server synthesizes the audio, and the audio is sent back.
Three things follow from where the work happens. With cloud TTS the text has to travel, a connection is required at the moment of speech, and every sentence costs the provider server time, which is why cloud readers count characters, minutes or hours. With local TTS the text stays where it is, the connection is irrelevant, and there is no running bill to recover.
Local no longer means the flat, robotic system voices: neural speech models are now small enough to fit on a phone and fast enough to run there in real time.
Why local TTS matters on a phone
- Privacy by architecture. Your books, PDFs and scanned pages are not uploaded, because nothing needs them anywhere else, and there is no account. The long version is on the private text to speech page.
- Airplane mode, tunnels, bad signal. Speech made on the device does not notice the network is gone. More under offline text to speech.
- No per-character meter. When the voice costs nothing to produce, there is no reason to ration it: no daily quota, no monthly hours, no premium voice tier.
- No queue. No upload and no rendering job: press play and the first sentence starts as soon as it is generated.
How Sandbook does local TTS
- The engine is Kokoro-82M, an open-weight neural text-to-speech model with 82 million parameters, run on the device with Apple's Core ML on the CPU, the same engine on iPhone, iPad and Mac (it ran on Apple MLX on the GPU before).
- Nothing to fetch. The model and all 41 voices come inside the app download, with no API key or activation step.
- Sentence by sentence. Each sentence is synthesized as you listen, at 24 kHz, so playback starts after the first sentence rather than after the whole chapter. Speed runs from 0.5× to 2.5×, and each word is highlighted as it is spoken.
- 41 voices in six languages: 28 in English (20 American, 8 British), 3 in Spanish, 1 in French, 2 in Italian, 3 in Brazilian Portuguese and 4 in Hindi.
- What it reads: EPUB, PDF, Word, TXT, RTF, Markdown and HTML files, web pages, camera scans and photos (recognized on the device), and typed or pasted text. A chapter or a whole book can be exported as one M4A file with chapter markers.
- In a desktop browser, too: try Kokoro TTS in a browser tab, where the speech is generated on your computer and nothing is uploaded.
What local TTS costs you, honestly
- A bigger download. The model and all 41 voices come inside the app download. Cloud apps can be smaller because their voices live elsewhere.
- A recent enough device. Sandbook needs iOS 18.2 on an iPhone XS, iPhone XR or later, or an iPad with an A12 chip or later. If you are not sure about yours, see text to speech on older iPhones.
- Fewer languages than cloud services. Six voice languages, not dozens. A server can hold a catalog no phone has room for.
How to check it is really local
Any app can say it runs on the device. The test takes a minute:
- Open a book, a PDF or some pasted text in the app.
- Turn on airplane mode, with Wi-Fi off as well.
- Press play and listen to a full chapter, then try a voice you have not used yet.
Speech generated on a server cannot survive that test. In Sandbook the only thing that waits for a connection is adding a new web page. Scanning a printed page with the camera works in airplane mode too, because the text recognition also runs on the device.
Questions about local TTS
Is local TTS as good as cloud TTS?
For reading a book aloud, the gap is much smaller than it used to be. Sandbook runs Kokoro-82M, a neural text-to-speech model, at 24 kHz on the iPhone itself, which is a different class of voice from the older robotic system voices that on-device speech once meant. Where cloud TTS still wins is breadth: a server can hold thousands of voices in dozens of languages, and a phone cannot. You can judge the quality yourself, because every one of Sandbook's 41 voices has a sample on the voices page.
Does local TTS work offline?
Yes, that is what makes it local. In Sandbook the model and all 41 voices ship inside the app, so EPUB, PDF, Word (.docx), TXT, RTF, Markdown and HTML files, camera scans and photos, and typed or pasted text all read aloud in airplane mode. The one thing that needs a connection is adding a new web page, because the page has to be fetched before it can be read; once saved, it reads offline too.
Which iPhones can run local TTS?
Sandbook needs iOS 18.2 or later on an iPhone XS, iPhone XR or later, or an iPad with an A12 chip or later, and its voices run on every one of them. An iPhone that cannot run iOS 18.2, like an iPhone X or an iPhone 8, can still use the text to speech built into iOS.
Is Sandbook's local TTS free?
Yes, completely. There is no in-app purchase, no subscription, no ads, no account and no character or minute limit. Because the speech is generated on your own device, reading a chapter costs nothing to anyone, so there is nothing to meter. All 41 voices in six languages are available from first launch.
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
Local TTS you can test in airplane mode
Install it, turn the radios off, and read a whole chapter. What you read never leaves your device. Free, no account, 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