Show HN: An improved kokoro distillation in 9MB

Full-stack developer MithrilMan released an improved distillation of the Kokoro text-to-speech model at roughly 9MB with about 25x real-time factor, publishing the Apache-2.0 code on GitHub and weights on Hugging Face under the project name teacher-in-the-loop. The developer said the original paper's training recipe was not replicable, so he researched and identified the cause and added tweaks, and extended the work beyond the paper's single English voice to Italian. The package installs via pip install teacher-in-the-loop, and the paper is published at DOI 10.5281/zenodo.23400040.

I'm a 47yo full stack developer that's investing all his free time and not only that into thinkering, building, researching with AI. Some days ago I found a paper about distilling kokoro on a ~9MB with around 25x RTF. It was very interesting and since the experiment was on a single english voice and I'm Italian, I wanted to see if I could manage to distill it in my language but the training recipe wasn't replicable. So I spent some days researching, experimenting and found the why and even some more tweaks to improve the overall result. The demo I linked shows you how close it is to the original kokoro and you can play straight in the browser If you use the package locally it's faster Paper: https://doi.org/10.5281/zenodo.23400040 https://doi.org/10.5281/zenodo.23400040 Code Apache-2.0 : https://github.com/MithrilMan/teacher-in-the-loop https://github.com/MithrilMan/teacher-in-the-loop Weights: https://huggingface.co/MithrilMan/teacher-in-the-loop https://huggingface.co/MithrilMan/teacher-in-the-loop pip install teacher-in-the-loop P.S. I wanted to publish the paper on arXiv but I need to be endorsed, would be nice if anyone that can help me could endorse me arxiv.org/auth/endorse?x=GPI7BZ Honestly would be just to be added to the hf papers, I don't seek fame nor fortune lol Comments URL: https://news.ycombinator.com/item?id=50110729 https://news.ycombinator.com/item?id=50110729 Points: 1 Comments: 0