I used over 3 minutes of clean audio clips from episodes of the anime and trained it on Google colab for 250 epochs.
Tip: Lower the transpose number and pitch for better accurate results, for a male to male voice between -6 and -12 is accurate.
Tags: Game Freak, Nintendo, Anime, huggingface, ai-model
Related URL: https://huggingface.co/Ryanham1lton/Croagunk/resolve/main/Croagunk.zip?download=true
Download Link: https://huggingface.co/Ryanham1lton/Croagunk/resolve/main/Croagunk.zip?download=true Hosted on Hugging Face