Description

Here's My Bones Model: 1000 Epoch & 110,000 Steps. I trained this with around an hour or more of audio samples. Should work okay with singing and speaking. Trained on his music predominantly but added one speaking sample to help it with regular speaking. Currently, the model has some background audio issues, but is completely useable at this stage. I may train on a cleaner dataset in the future, this was mostly just my first test model.