First time using a lower learning rate for training. It paid off and I plan to use it more for small datasets such as this.
Australian audio is recommended for obvious reasons, otherwise replicating his accent wouldn't be easy. I also recommend an index ratio of 0.4-0.6.
Interested in commissioning high quality models? Feel free to visit my shop! https://discord.com/channels/1159260121998827560/1200536399850524883
DOWNLOAD: https://huggingface.co/ScruffyRVC/RFCRVCV2/resolve/main/mammon_443e_4873s.zip?download=true