Training models requires a few things.
1) Dataset, the cleaner the better. A few good guide for this are https://docs.google.com/document/d/13ebnzmeEBc6uzYCMt-QVFQk-whVrK4zw8k7_Lw3Bv_A/edit?usp=sharing and https://rentry.org/RVC_making-models
2) Training algorithm, in this guide we will use RVC Disconnected https://colab.research.google.com/drive/1XIPCP9ken63S7M6b5ui1b36Cs17sP-NS
3) Compute, this is also handled with RVC Disconnected
Hopefully I will have to make a more comprehensive guide soon