Read text aloud with natural voices — in your browser
Paste or open text and listen. The voices run on your computer; nothing you read is uploaded.
Paste or open text and listen. The voices run on your computer; nothing you read is uploaded.
Most natural-sounding text-to-speech tools send your text to a server and play back what comes out. Whatever you were reading — an email, a contract, your own writing — passes through someone else's computer. SpeakFit runs its voices inside your browser instead, using the open Kokoro voice model. Your text is read where it already is. It is never uploaded.
The voices are a one-time download of about 100 MB. You press "Download the voices" once and they are saved in your browser's own storage. After that, SpeakFit works offline.
No. The voices run on your own device. The only thing SpeakFit downloads is the voice files, once, when you press "Download the voices". That request carries no information about you or your text.
English — American and British voices. The voices can read other languages' words, but with an English accent.
SpeakFit is free with four voices, for any length of text. A one-time purchase adds twelve more voices and saving speech as an audio file. There is no subscription.
Yes. SpeakFit is free software under the GNU GPL, version 3 or later — get the source code.
Step-by-step instructions · Privacy policy
Every tool in the suite runs entirely on your own device — same approach, different job. Each is also available as a browser extension.