kale-tts-1 released open-weight in base/tiny/nano sizes, dataset included
NERDDISCO · x · 2026-11-20
@NERDDISCO has released kale-tts-1, their first speech synthesis model on Hugging Face, with open weights in three sizes (base, tiny, nano) plus the training dataset and an online demo. Integrations with mlx-audio, llama.cpp, and audio.cpp are next — all three sizes already run locally, with pull requests coming soon.
Related event: NERDDISCO releases kale-tts-1, a 14.8MB open TTS family(7 posts)→