Text to Speech
Info
- Powered by Kokoro-82M by hexgrad, built on StyleTTS 2 by yl4579. Running locally in your browser via kokoro-js and Transformers.js.
- Your data never leaves your browser.
Warning
- Using this app (adding an image or images) will download ~90MBs of AI onto your device! Plus the voices!
- It also uses a fair bit of processing power!
The quality of this model is dramatically worse than the state of the art for large models, but considering it is on device, and is unlimited and free, it’s still quite impressive.
In-browser models will likely improve significantly due to optimization opportunities and expanding WebGPU support.
Ready. Model will download (~92 MB) when you click Generate.