Kokoro-82M Enables Local Text-to-Speech

amos_gyamfi · x · 2026-07-14

Built on StyleTTS2, Kokoro-82M is an on-device text-to-speech model with 82 million parameters that supports 24kHz audio output.

The creator notes it can run on iOS 27 and macOS 27 via Core AI, providing links to Hugging Face and the Core AI Model Zoo.

Original post →

More from Multimodal

Multimodal channel →