Do we have to write code for preprocessing the dataset?

[ashish-roopan]

I cant find any info on the repo about preprocessing the input data before training.
Can you please say which dataset to use and how to train the model with that dataset?

[astorfi]

Please check https://github.com/astorfi/lip-reading-deeplearning/blob/master/code/speech-input/input_feature.py and the paper for further details.