Viewing profile — mrfakename

Cool, it looks like this model is pretty similar to StyleTTS 2? Would it be possible to confirm?

Hi HN, I just launched TTS Arena V2 - a platform for benchmarking TTS models by blind A/B testing. The goal is to make it easy to compare quality between open-source and commercial…

Wow! Impressive. How did they manage to fit GPT-2 in a spreadsheet? A question for the author: will it be open sourced?

Looks like the website (amazon-ltts-paper.com) now redirects to amazon.science. They took out the "Ethical Statement" section. (The original page can still be accessed from the Way…

I agree. I also think the Llama naming was confusing - versioning by capitalization? (LLaMA vs Llama)

Have you considered open source LLMs?

Note that it's actually "Mistral Next" not "Mixtral Next" - so it isn't necessarily a MoE. For example, an early version of Mistral Medium (Miqu) was not a MoE but instead a Llama …

Sadly they didn't release the code or models