This repo provides the network code and the processed samples of the manuscript "Glance and Gaze: A Collaborative Learning Framework for Single-channel Speech Enhancement", which was accepted by Elsevier Applied Acoustics and can be accessed here. You can train the network with your own pipeline. As I have rearranged the code, if you find any bug during the training process, please feel free to inform me.