About ListenKit

ListenKit reads text aloud in your browser. The natural voice is Kokoro-82M, an open (Apache-2.0) speech model that runs on your own device through WebGPU or WebAssembly using kokoro-js and transformers.js. The model files (about 90 MB) are downloaded once from Hugging Face and cached by your browser. A System voice option uses the speech voices built into your browser or operating system.

Honest limits: there are eight natural English voices (far fewer than paid apps), scanned PDFs without a text layer are not supported (no OCR), and slower devices can pause between sentences with the natural voice.

ListenKit is made by RT Technologies Inc., a small Canadian software company that builds free web tools supported by optional tips. There is no paid plan and tips never unlock features.

Bug reports and feature requests: hello@rt-tech.ca.