This repo provides the processed samples on the noisy audios of the VoiceBank_DEMAND noisy testet, which are enhanced by our FDFNet model from the manuscript "A Two-stage Framework in Cross-spectrum Domain for Real-time Speech Enhancement". The source code will be available later.