Discriminative Enhancement for Single Channel Audio Source Separation using Deep Neural Networks

TL;DR

This paper introduces a discriminative enhancement approach using deep neural networks to improve the quality of single channel audio source separation by reducing distortion and interference among separated sources.

Contribution

It proposes a novel DNN-based discriminative enhancement method that jointly enhances all separated sources to minimize residual interference, improving separation quality.

Findings

Discriminative enhancement reduces distortion in separated sources.

The method decreases interference between sources.

Experimental results show improved separation quality.

Abstract

The sources separated by most single channel audio source separation techniques are usually distorted and each separated source contains residual signals from the other sources. To tackle this problem, we propose to enhance the separated sources to decrease the distortion and interference between the separated sources using deep neural networks (DNNs). Two different DNNs are used in this work. The first DNN is used to separate the sources from the mixed signal. The second DNN is used to enhance the separated signals. To consider the interactions between the separated sources, we propose to use a single DNN to enhance all the separated sources together. To reduce the residual signals of one source from the other separated sources (interference), we train the DNN for enhancement discriminatively to maximize the dissimilarity between the predicted sources. The experimental results show that using discriminative enhancement decreases the distortion and interference between the separated sources.