/vt/ - Virtual Youtubers

You might be interested in this nyan anon if you have a somewhat decent graphics card. (at least a 1060)
>Download RVC GUI:
https://github.com/Tiger14n/RVC-GUI/releases/tag/Windows-pkg
>Download this AI nyanners zip file:
https://huggingface.co/legitdark/Nyanners-By-Dan/resolve/main/Nyanners-By-Dan.zip
>Click import model from .zip in RVC GUI:
select the downloaded AI nyanners zip file
>find a TTS service that lets you download audio files I use https://ttsmp3.com/ because it's free
Post your LLM AI generations into the TTS service and download the audio
>select the nyanners model in RVC GUI:
Use the harvest method and adjust the pitch until the outputted voice sounds correct. (You won't have to adjust it again if you keep using the same TTS voice as the input.)
Here's an example output: https://voca.ro/13D38tH9CBk2
You can get much better results if you use a better TTS service or your own voice as an input. I can do this pretty much instantly with a 1070. I also found a bunch of other custom vtuber AI models in the AI Hub discord. You can download and use them in the same way as the nyanners model, but the nyanners model is really the only one I care about.