Fine-tuning trOCR on handwritten Sinhala text

Hi @NielsRogge,
I wanted to fine tune trOCR model on extracting the handwritten sinhala language text. I really don't know how to start. Would you please guide something on this. Is there pretrained model on this? Or should I try it from scratch? Would you please, share some material or repository on this?

Your guidence would be much appreciated.