Single channel voice separation for unknown number of speakers under reverberant and noisy settings

Shlomo E. Chazan; Lior Wolf; Eliya Nachmani; Yossi Adi

doi:10.1109/ICASSP39728.2021.9413627

Single channel voice separation for unknown number of speakers under reverberant and noisy settings

Shlomo E. Chazan, Lior Wolf, Eliya Nachmani, Yossi Adi

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

22 Scopus citations

Abstract

We present a unified network for voice separation of an unknown number of speakers. The proposed approach is composed of several separation heads optimized together with a speaker classification branch. The separation is carried out in the time domain, together with parameter sharing between all separation heads. The classification branch estimates the number of speakers while each head is specialized in separating a different number of speakers. We evaluate the proposed model under both clean and noisy reverberant settings. Results suggest that the proposed approach is superior to the baseline model by a significant margin. Additionally, we present a new noisy and reverberant dataset of up to five different speakers speaking simultaneously.

Original language	English
Title of host publication	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	3730-3734
Number of pages	5
ISBN (Electronic)	9781728176055
DOIs	https://doi.org/10.1109/ICASSP39728.2021.9413627
State	Published - 2021
Externally published	Yes
Event	2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021 - Virtual, Toronto, Canada Duration: 6 Jun 2021 → 11 Jun 2021

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2021-June
ISSN (Print)	1520-6149

Conference

Conference	2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021
Country/Territory	Canada
City	Virtual, Toronto
Period	6/06/21 → 11/06/21

Bibliographical note

Publisher Copyright:
© 2021 IEEE

Keywords

Source separation
Speaker classification
Speech processing

Access to Document

10.1109/ICASSP39728.2021.9413627

Cite this

Chazan, S. E., Wolf, L., Nachmani, E., & Adi, Y. (2021). Single channel voice separation for unknown number of speakers under reverberant and noisy settings. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 3730-3734). (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2021-June). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP39728.2021.9413627

Chazan, Shlomo E. ; Wolf, Lior ; Nachmani, Eliya et al. / Single channel voice separation for unknown number of speakers under reverberant and noisy settings. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 3730-3734 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{1172550cf3bd4b49a1d94b2d0dc976f6,

title = "Single channel voice separation for unknown number of speakers under reverberant and noisy settings",

abstract = "We present a unified network for voice separation of an unknown number of speakers. The proposed approach is composed of several separation heads optimized together with a speaker classification branch. The separation is carried out in the time domain, together with parameter sharing between all separation heads. The classification branch estimates the number of speakers while each head is specialized in separating a different number of speakers. We evaluate the proposed model under both clean and noisy reverberant settings. Results suggest that the proposed approach is superior to the baseline model by a significant margin. Additionally, we present a new noisy and reverberant dataset of up to five different speakers speaking simultaneously.",

keywords = "Source separation, Speaker classification, Speech processing",

author = "Chazan, {Shlomo E.} and Lior Wolf and Eliya Nachmani and Yossi Adi",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE; 2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021 ; Conference date: 06-06-2021 Through 11-06-2021",

year = "2021",

doi = "10.1109/ICASSP39728.2021.9413627",

language = "אנגלית",

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

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

pages = "3730--3734",

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

address = "ארצות הברית",

}

Chazan, SE, Wolf, L, Nachmani, E & Adi, Y 2021, Single channel voice separation for unknown number of speakers under reverberant and noisy settings. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2021-June, Institute of Electrical and Electronics Engineers Inc., pp. 3730-3734, 2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021, Virtual, Toronto, Canada, 6/06/21. https://doi.org/10.1109/ICASSP39728.2021.9413627

Single channel voice separation for unknown number of speakers under reverberant and noisy settings. / Chazan, Shlomo E.; Wolf, Lior; Nachmani, Eliya et al.
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2021. p. 3730-3734 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2021-June).

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

TY - GEN

T1 - Single channel voice separation for unknown number of speakers under reverberant and noisy settings

AU - Chazan, Shlomo E.

AU - Wolf, Lior

AU - Nachmani, Eliya

AU - Adi, Yossi

PY - 2021

Y1 - 2021

N2 - We present a unified network for voice separation of an unknown number of speakers. The proposed approach is composed of several separation heads optimized together with a speaker classification branch. The separation is carried out in the time domain, together with parameter sharing between all separation heads. The classification branch estimates the number of speakers while each head is specialized in separating a different number of speakers. We evaluate the proposed model under both clean and noisy reverberant settings. Results suggest that the proposed approach is superior to the baseline model by a significant margin. Additionally, we present a new noisy and reverberant dataset of up to five different speakers speaking simultaneously.

AB - We present a unified network for voice separation of an unknown number of speakers. The proposed approach is composed of several separation heads optimized together with a speaker classification branch. The separation is carried out in the time domain, together with parameter sharing between all separation heads. The classification branch estimates the number of speakers while each head is specialized in separating a different number of speakers. We evaluate the proposed model under both clean and noisy reverberant settings. Results suggest that the proposed approach is superior to the baseline model by a significant margin. Additionally, we present a new noisy and reverberant dataset of up to five different speakers speaking simultaneously.

KW - Source separation

KW - Speaker classification

KW - Speech processing

UR - http://www.scopus.com/inward/record.url?scp=85109484439&partnerID=8YFLogxK

U2 - 10.1109/ICASSP39728.2021.9413627

DO - 10.1109/ICASSP39728.2021.9413627

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

AN - SCOPUS:85109484439

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

SP - 3730

EP - 3734

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

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021

Y2 - 6 June 2021 through 11 June 2021

ER -

Chazan SE, Wolf L, Nachmani E, Adi Y. Single channel voice separation for unknown number of speakers under reverberant and noisy settings. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2021. p. 3730-3734. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP39728.2021.9413627

Single channel voice separation for unknown number of speakers under reverberant and noisy settings

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this