FunASR
FunASR copied to clipboard
Finetube the punctuation model
Is it possible to finetube the punctuation model independently? What is the format of the labels of the training data?