Speech Dereverberation Using Fully Convolutional Networks

TL;DR

This paper explores the use of fully convolutional networks, including U-Net and GAN architectures, for speech dereverberation from single-microphone recordings, demonstrating superior performance over existing methods.

Contribution

It introduces novel FCN-based models for speech dereverberation, applying image processing techniques to audio spectrograms, and evaluates their effectiveness on standard datasets.

Findings

Our method outperforms existing approaches in most cases.

U-Net and GAN architectures improve dereverberation quality.

The approach leverages image processing techniques for audio enhancement.

Abstract

Speech derverberation using a single microphone is addressed in this paper. Motivated by the recent success of the fully convolutional networks (FCN) in many image processing applications, we investigate their applicability to enhance the speech signal represented by short-time Fourier transform (STFT) images. We present two variations: a "U-Net" which is an encoder-decoder network with skip connections and a generative adversarial network (GAN) with U-Net as generator, which yields a more intuitive cost function for training. To evaluate our method we used the data from the REVERB challenge, and compared our results to other methods under the same conditions. We have found that our method outperforms the competing methods in most cases.