RaD-Net: A Repairing and Denoising Network for Speech Signal Improvement

This paper introduces our repairing and denoising network (RaD-Net) for theICASSP 2024 Speech Signal Improvement (SSI) Challenge. We extend our previousframework based on a two-stage network and propose an upgraded model.Specifically, we replace the repairing network with COM-Net from TEA-PSE. Inaddition, multi-resolution discriminators and multi-band discriminators areadopted in the training stage. Finally, we use a three-step training strategyto optimize our model. We submit two models with different sets of parametersto meet the RTF requirement of the two tracks. According to the officialresults, the proposed systems rank 2nd in track 1 and 3rd in track 2.