I'm trying to pretrain a model with another langauge

[sympathize123]

I read the pretrain section however it seems that we need to prepare the dataset in LRS3 format but the data I have is video clip of people talking and wav audio file, and the time for each sentence ends. Is there a function or a file i can preprocess into your dataset format?

[chevalierNoir]

There is no single command for data preparation yet. You can refer to this thread for details on how to prepare your own dataset and train a model.