Trained with mangio crepe 8 hop length with an 11 minute dataset of stems and acoustic covers. The stems were not super high quality, but this was the best that I could do after around two months of trying to get it to sound good. This model is not meant for speaking, but is much better at singing. Please credit me if you use this model.

Additional Details:

Tags: No tags available

Download Link: https://huggingface.co/TheGreasyGamer/DustinBates/resolve/main/DustinBates.zip?download=true Hosted on Hugging Face