Five days ago, I published a blog post, describing why TorToiSe could not be fine-tuned.

Today, I have released a fork of DL-Art-School with TorToiSe fine-tuning code. How did that happen?

TorToiSe 🐢 is an open-source Text-To-Speech (TTS) neural network that creates fairly authentic & realistic voices. Checkpoints for local inference have been available since April last year, but its users are seemingly unable to fine-tune the model with additional voice data.

Why is this the case, and how could it be fixed?

On 9th Feb, the National University of Singapore implemented broad restrictions against the use of AI tools to generate work:

I made a fork of TorToiSe with much faster inference speed. Here are the summarised results:

TL;DR

• GPT-NeoX-20B appears to be anisotropic (Isotropy: 0.197)
• Int8 quantisation appears to have ~no effect on isotropy
• I am new to ML so the above could be false, feel free to poke at my findings here