Implementation of paper "DPCRN: Dual-Path Convolution Recurrent Network for Single Channel Speech Enhancement". This work got the third place in Deep Noise Suppression Challenge.
Requirements: tensorflow>=1.14, numpy, matplotlib, librosa, sondfile.
Note that the real-time inference can only run on the tensorflow=1.x.
The final results on the blind test set of DNS3 is available on https://github.com/Le-Xiaohuai-speech/DPCRN_DNS3_Results.
run python main.py to test the model on a single file.