Using custom data for model training and testing

I need to train and evaluate the ConvTasNet model on speech enhancement for a custom dataset.
How should I go about formatting the data, and what changes need to be made in the run.sh file.

Start by locating the ConvTasNet training and evaluation entry points and the run.sh file. Inspect how the existing dataset is formatted and configured, then determine what guidance is needed for a custom speech-enhancement dataset. Done means the required data format and run.sh changes are documented clearly enough to train and evaluate the model.

Written by the indexing model from the issue text.