ModelKokoro
Kokoro 82M
A lightweight neural network text-to-speech (TTS) model with only 82 million parameters. Despite its small size, the model produces relatively natural-sounding speech and is designed for local deployment on edge devices.
Timeline (last 90 days)
Oct 12
TrailLens Audio uses the Kokoro 82M local neural network text-to-speech model to provide voice narration without relying on paid voice services.
Unverified50%
Oct 12
Kokoro 82M is a neural network TTS model with only 82 million parameters.
Unverified50%