Nano-Edge-TTS – 70x realtime, Lightweight Kokoro-based TTS model
Comments
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Summary
Nano-Edge-TTS, released by Mike Chumba, replaces the 53.3M-parameter waveform decoder of the open Kokoro-82M text-to-speech model with a 1.75M-parameter harmonic-plus-noise decoder that runs no neural network at audio rate.