Speech dereverberation using fully convolutional networks

Ori Ernst, Shlomo E. Chazan, Sharon Gannot, Jacob Goldberger

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

53 Scopus citations

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.

Original languageEnglish
Title of host publication2018 26th European Signal Processing Conference, EUSIPCO 2018
PublisherEuropean Signal Processing Conference, EUSIPCO
Pages390-394
Number of pages5
ISBN (Electronic)9789082797015
DOIs
StatePublished - 29 Nov 2018
Event26th European Signal Processing Conference, EUSIPCO 2018 - Rome, Italy
Duration: 3 Sep 20187 Sep 2018

Publication series

NameEuropean Signal Processing Conference
Volume2018-September
ISSN (Print)2219-5491

Conference

Conference26th European Signal Processing Conference, EUSIPCO 2018
Country/TerritoryItaly
CityRome
Period3/09/187/09/18

Bibliographical note

Publisher Copyright:
© EURASIP 2018.

Fingerprint

Dive into the research topics of 'Speech dereverberation using fully convolutional networks'. Together they form a unique fingerprint.

Cite this