MATLAB-denoise

MathWorks-Excellence-in-Innovation project 193 & SJTU EE397 project

Introduction

This is the course project of SJTU EE397 course and project 193 of MathWorks Excellence in Innovation Projects. We learn from the idea of RNNoise [1] and implement a speech noise reduction system based on MATLAB deep learning.

Remind

• The TrainNetworks folder is used to train and simply test the network, use the DenoiseTrain.m to start it.
• The code in objective comment is used to evalute the effect, with the objective comment.
• Before run it, please download the voicebox in your MATLAB from http://www.ee.ic.ac.uk/hp/staff/dmb/voicebox/voicebox.html

References

[1] Valin J M. A hybrid DSP/deep learning approach to real-time full-band speech enhancement[C]//2018 IEEE 20th international workshop on multimedia signal processing (MMSP). IEEE, 2018: 1-5.