Home
Author Guide
Editor Guide
Reviewer Guide
Special Issues
Special Issue Introduction
Special Issues List
Topics
Published Issues
2024
2023
2022
2021
2020
2019
2018
2017
2016
2015
2014
2013
2012
2010
2009
2008
2007
2006
journal menu
Aims and Scope
Editorial Board
Indexing Service
Article Processing Charge
Open Access Policy
Publication Ethics
Digital Preservation Policy
Editorial Process
Subscription
Contact Us
General Information
ISSN:
1796-2021 (Online); 2374-4367 (Print)
Abbreviated Title:
J. Commun.
Frequency:
Monthly
DOI:
10.12720/jcm
Abstracting/Indexing:
Scopus
;
DBLP
;
CrossRef
,
EBSCO
,
Google Scholar
;
CNKI,
etc.
E-mail questions
or comments to
editor@jocm.us
Acceptance Rate:
27%
APC:
800 USD
Average Days to Accept:
88 days
3.4
2023
CiteScore
51st percentile
Powered by
Article Metrics in Dimensions
Editor-in-Chief
Prof. Maode Ma
College of Engineering, Qatar University, Doha, Qatar
I'm very happy and honored to take on the position of editor-in-chief of JCM, which is a high-quality journal with potential and I'll try my every effort to bring JCM to a next level...
[Read More]
What's New
2024-08-20
Vol. 19, No. 8 has been published online!
2024-07-22
Vol. 19, No. 7 has been published online!
2024-06-20
Volume 19, No. 4 has been indexed by Scopus.
Home
>
Published Issues
>
2020
>
Volume 15, No. 11, November 2020
>
Speech Separation in the Frequency Domain with Autoencoder
Hao D. Do
1,2,3
, Son T. Tran
1,2
, and Duc T. Chau
1,2
1. University of Science, Ho Chi Minh City, Vietnam
2. Vietnam National University, Ho Chi Minh City, Vietnam
3. OLLI Technology JSC, Ho Chi Minh City, Vietnam
Abstract
—Speech separation plays an important role in a speech-related system because it can denoise, extract and enhance speech signal, and after all improve the accuracy and performance of the system. In recent years, many approaches only separate the speech out of commonly high-frequency noise or a particular background sound. We propose a more powerful approach, combining an autoencoder and a bandpass filter to separate speech signals. This combination can extract the speech in the mixture with not only high-frequency noise but also many kinds of different background sounds. Our approach can be flexibly applied for the new background sounds. Experimental results show that our model can extract fastly and effectively the speech signal with 9.01 dB in SIR and 11.26 in SDR. On the other hand, we can adjust the passband to identify the range of frequency at the output signal to apply for particular applications.
Index Terms
—Speech separation, autoencoder, bandpass filter, frequency domain
Cite: Hao D. Do, Son T. Tran, and Duc T. Chau, "Speech Separation in the Frequency Domain with Autoencoder," Journal of Communications vol. 15, no. 11, pp. 841-848, November 2020. Doi: 10.12720/jcm.15.11.841-848
Copyright © 2020 by the authors. This is an open access article distributed under the Creative Commons Attribution License (
CC BY-NC-ND 4.0
), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.
8-RC023
PREVIOUS PAPER
MOSFET Based Absorber for Reflected Signal in 5G Massive MIMO Base Station - A Circuit Perspective
NEXT PAPER
Last page