Log In Sign Up

MetricGAN+/-: Increasing Robustness of Noise Reduction on Unseen Data

by   George Close, et al.

Training of speech enhancement systems often does not incorporate knowledge of human perception and thus can lead to unnatural sounding results. Incorporating psychoacoustically motivated speech perception metrics as part of model training via a predictor network has recently gained interest. However, the performance of such predictors is limited by the distribution of metric scores that appear in the training data. In this work, we propose MetricGAN+/- (an extension of MetricGAN+, one such metric-motivated system) which introduces an additional network - a "de-generator" which attempts to improve the robustness of the prediction network (and by extension of the generator) by ensuring observation of a wider range of metric scores in training. Experimental results on the VoiceBank-DEMAND dataset show relative improvement in PESQ score of 3.8 generalisation to unseen noise and speech.


MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement

The discrepancy between the cost function used for training a speech enh...

Language and Noise Transfer in Speech Enhancement Generative Adversarial Network

Speech enhancement deep learning systems usually require large amounts o...

MetricGAN: Generative Adversarial Networks based Black-box Metric Scores Optimization for Speech Enhancement

Adversarial loss in a conditional generative adversarial network (GAN) i...

Variational Autoencoder for Speech Enhancement with a Noise-Aware Encoder

Recently, a generative variational autoencoder (VAE) has been proposed f...

SEGAN: Speech Enhancement Generative Adversarial Network

Current speech enhancement techniques operate on the spectral domain and...