OFF-THE-SHELF DEEP INTEGRATION FOR RESIDUAL-ECHO SUPPRESSION

Amir Ivry; Israel Cohen; Baruch Berdugo

doi:10.1109/ICASSP43922.2022.9747511

OFF-THE-SHELF DEEP INTEGRATION FOR RESIDUAL-ECHO SUPPRESSION

Amir Ivry, Israel Cohen, Baruch Berdugo

Technion-Israel Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

Residual-echo suppression (RES) systems suppress the echo and preserve the speech from a mixture of the two. In handsfree speech communication, RES may also be addressed as a source separation (SS) or speech enhancement (SE) problem, where the echo can be manipulated as an interfering speech signal. In this study, we fine-tune three pre-trained deep learning-based systems originally designed for RES, SS, and SE, and show that the best performing system for the task of RES varies with respect to the acoustic conditions. Then, we propose a real-time data-driven integration of these systems, where a neural network continuously tracks the system that achieves the best performance during both single-talk and double-talk periods. Experiments with 100 h of real and synthetic data show that the integrated system outperforms each individual system in terms of echo suppression and speech distortion in various acoustic environments.

Original language	English
Title of host publication	2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	746-750
Number of pages	5
ISBN (Electronic)	9781665405409
DOIs	https://doi.org/10.1109/ICASSP43922.2022.9747511
State	Published - 2022
Externally published	Yes
Event	47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Virtual, Online, Singapore Duration: 23 May 2022 → 27 May 2022

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2022-May
ISSN (Print)	1520-6149

Conference

Conference	47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022
Country/Territory	Singapore
City	Virtual, Online
Period	23/05/22 → 27/05/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE

Funding

This research was supported by the Pazy Research Foundation.

Funders	Funder number
Pazy Research Foundation

Keywords

Acoustic echo cancellation
deep learning
residual-echo suppression
speech enhancement
speech separation

Access to Document

10.1109/ICASSP43922.2022.9747511

Cite this

Ivry, A., Cohen, I., & Berdugo, B. (2022). OFF-THE-SHELF DEEP INTEGRATION FOR RESIDUAL-ECHO SUPPRESSION. In 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings (pp. 746-750). (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2022-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP43922.2022.9747511

Ivry, Amir ; Cohen, Israel ; Berdugo, Baruch. / OFF-THE-SHELF DEEP INTEGRATION FOR RESIDUAL-ECHO SUPPRESSION. 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 746-750 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{3dd84855fb28493b9a79d79f2988d52a,

title = "OFF-THE-SHELF DEEP INTEGRATION FOR RESIDUAL-ECHO SUPPRESSION",

abstract = "Residual-echo suppression (RES) systems suppress the echo and preserve the speech from a mixture of the two. In handsfree speech communication, RES may also be addressed as a source separation (SS) or speech enhancement (SE) problem, where the echo can be manipulated as an interfering speech signal. In this study, we fine-tune three pre-trained deep learning-based systems originally designed for RES, SS, and SE, and show that the best performing system for the task of RES varies with respect to the acoustic conditions. Then, we propose a real-time data-driven integration of these systems, where a neural network continuously tracks the system that achieves the best performance during both single-talk and double-talk periods. Experiments with 100 h of real and synthetic data show that the integrated system outperforms each individual system in terms of echo suppression and speech distortion in various acoustic environments.",

keywords = "Acoustic echo cancellation, deep learning, residual-echo suppression, speech enhancement, speech separation",

author = "Amir Ivry and Israel Cohen and Baruch Berdugo",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE; 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 ; Conference date: 23-05-2022 Through 27-05-2022",

year = "2022",

doi = "10.1109/ICASSP43922.2022.9747511",

language = "אנגלית",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "746--750",

booktitle = "2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings",

address = "ארצות הברית",

}

Ivry, A, Cohen, I & Berdugo, B 2022, OFF-THE-SHELF DEEP INTEGRATION FOR RESIDUAL-ECHO SUPPRESSION. in 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2022-May, Institute of Electrical and Electronics Engineers Inc., pp. 746-750, 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022, Virtual, Online, Singapore, 23/05/22. https://doi.org/10.1109/ICASSP43922.2022.9747511

OFF-THE-SHELF DEEP INTEGRATION FOR RESIDUAL-ECHO SUPPRESSION. / Ivry, Amir; Cohen, Israel; Berdugo, Baruch.
2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. p. 746-750 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2022-May).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - OFF-THE-SHELF DEEP INTEGRATION FOR RESIDUAL-ECHO SUPPRESSION

AU - Ivry, Amir

AU - Cohen, Israel

AU - Berdugo, Baruch

PY - 2022

Y1 - 2022

N2 - Residual-echo suppression (RES) systems suppress the echo and preserve the speech from a mixture of the two. In handsfree speech communication, RES may also be addressed as a source separation (SS) or speech enhancement (SE) problem, where the echo can be manipulated as an interfering speech signal. In this study, we fine-tune three pre-trained deep learning-based systems originally designed for RES, SS, and SE, and show that the best performing system for the task of RES varies with respect to the acoustic conditions. Then, we propose a real-time data-driven integration of these systems, where a neural network continuously tracks the system that achieves the best performance during both single-talk and double-talk periods. Experiments with 100 h of real and synthetic data show that the integrated system outperforms each individual system in terms of echo suppression and speech distortion in various acoustic environments.

AB - Residual-echo suppression (RES) systems suppress the echo and preserve the speech from a mixture of the two. In handsfree speech communication, RES may also be addressed as a source separation (SS) or speech enhancement (SE) problem, where the echo can be manipulated as an interfering speech signal. In this study, we fine-tune three pre-trained deep learning-based systems originally designed for RES, SS, and SE, and show that the best performing system for the task of RES varies with respect to the acoustic conditions. Then, we propose a real-time data-driven integration of these systems, where a neural network continuously tracks the system that achieves the best performance during both single-talk and double-talk periods. Experiments with 100 h of real and synthetic data show that the integrated system outperforms each individual system in terms of echo suppression and speech distortion in various acoustic environments.

KW - Acoustic echo cancellation

KW - deep learning

KW - residual-echo suppression

KW - speech enhancement

KW - speech separation

UR - http://www.scopus.com/inward/record.url?scp=85131251009&partnerID=8YFLogxK

U2 - 10.1109/ICASSP43922.2022.9747511

DO - 10.1109/ICASSP43922.2022.9747511

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85131251009

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 746

EP - 750

BT - 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022

Y2 - 23 May 2022 through 27 May 2022

ER -

Ivry A, Cohen I, Berdugo B. OFF-THE-SHELF DEEP INTEGRATION FOR RESIDUAL-ECHO SUPPRESSION. In 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2022. p. 746-750. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP43922.2022.9747511

OFF-THE-SHELF DEEP INTEGRATION FOR RESIDUAL-ECHO SUPPRESSION

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this