Here we will develop a general model for Speech Separation.Without any dependency on wsj0 dataset.We will proceed from scratch.And we will be able to do speech separation for any dataset
The first script will move ur data into folders wsj0-hin/tr , wsj0-hin/cv and wsj0-hin/tt. Make sure you create these three folders manually before running the script.
An example of sample file as a result of the script is 'mixhin_2_spk_tt.txt'