Kokoro TTS online, free and private in your browser

The model runs on your own computer, plays the first sentence while the rest renders, and gives you a WAV to download. When you press Start, your browser downloads the Kokoro-82M model once and keeps it in its cache. From then on everything happens in this tab. Your text is split into sentences, each sentence is turned into phonemes, and the model turns the phonemes into 24 kHz audio in the voice you picked. The first sentence starts playing as soon as it is ready, while the next ones render behind it.

English text goes through Kokoro's own English phonemizer, which also spells out numbers, prices and times. Spanish, French, Italian, Portuguese and Hindi go through eSpeak NG, the same phonemizer Kokoro uses for those languages, loaded only when you pick one of those voices. The inference runs in a background worker, so the page stays responsive.

The tool picks the engine for you. There is no quality setting to get wrong, because the smaller quantized models produce garbled speech on WebGPU.

• WebGPU (graphics card): the full-precision model, a one-time 326 MB download. On a recent laptop, a 40-word paragraph takes about 3 seconds and the first sentence plays after about 1 second. Current Chrome and Edge on a computer have WebGPU; so does Safari 26 on a Mac.
• CPU (WebAssembly): the 8-bit model, a one-time 92 MB download, used when WebGPU is not available. It is slower than real time: expect the first sentence after 10 to 15 seconds and 20 to 40 seconds for the same paragraph, depending on your processor.

Either way, the second visit skips the download: the model loads from the browser cache in a few seconds. Plan on 1 to 1.5 GB of memory in the tab while it runs.

The voice decides the language. The tool offers the same 41 voices as the Sandbook app, with the app's names and descriptions:

• English: 28 voices, 20 American and 8 British, including Heart, Bella, Michael, Emma and George.
• Spanish: 3 voices (Dora, Alex, Santa).
• French: 1 voice (Siwis).
• Italian: 2 voices (Sara, Nicola).
• Brazilian Portuguese: 3 voices (Dora, Alex, Santa).
• Hindi: 4 voices (Alpha, Beta, Omega, Psi).

To audition a voice without loading anything, press the play button next to the voice picker, or listen to all the Kokoro voice samples. Speed runs from 0.5× to 2.0×, applied inside the model, so a fast voice still sounds like a person talking quickly.

Download. Every generation can be saved as a 24 kHz mono WAV file. MP3 and M4A export are not available yet.

Usage rights. Kokoro-82M is published under the Apache-2.0 license, which allows commercial use. The audio you generate is yours; Sandbook claims nothing over it.

Limits. Up to 1,500 characters per generation, about a minute and a half of speech. The model needs a desktop-class amount of memory, so the tool is meant for computers. Phones may not have the memory for the browser version, and on a phone the page shows voice samples before offering to load anything. Closing the tab right after the first download can occasionally stop the browser from finishing its cache write, in which case the next visit downloads the model again.

Privacy. Everything stays in your browser. Your text and the audio never leave this tab; the only files the tool fetches are the model, the voice you pick and, for non-English voices, the phonemizer. The page counts how often the tool is used, never what you type.

This page is a demo of the model. Sandbook is the app: it runs the same Kokoro-82M model on iPhone and iPad through Apple MLX, fully offline, with all 41 voices and no length limit. It reads EPUB books and PDFs, web pages and scanned pages aloud, highlights each word as it is spoken, and remembers where you stopped. Free, with no account and nothing to buy. It needs an iPhone 12 or later on iOS 18.2.

How the model runs on a phone is explained on the Kokoro TTS app for iPhone page. If you only need a phone to read the screen aloud, iOS already can: text to speech on iPhone with the built-in settings. And for how Sandbook compares with Speechify, NaturalReader and the other readers, see the iPhone reader alternatives.