FreeTTS

Studio

Real AI speech, synthesized on your device. Nothing you type is sent anywhere.

Kokoro-82M — high-quality neural voices, 28 verified English voices. The model (~92 MB) downloads once, on first use, then works offline.

Text to speak

0 / 5,000

Voice

Loading voices…

Speaking speed

1.00×

0.5× slower to 2× faster.

No accounts, no tracking, no analytics. How privacy works