Models / hexgrad / Kokoro 82M

Kokoro 82M

JSON →
audio

An 82-million-parameter text-to-speech model optimized for fast and natural speech synthesis.

audiostreaming
homepage ↗
Specs
Context & limits
—
context window
—
max output
Lifecycle
Dates
released2024-09-01
Resources
homepagehttps://huggingface.co/hexgrad/Kokoro-82M ↗