Sudo RM -RF: Efficient networks for universal audio source separation

Efthymios Tzinis; Zhepei Wang; Paris Smaragdis

doi:10.1109/MLSP49062.2020.9231900

Sudo RM -RF: Efficient networks for universal audio source separation

Efthymios Tzinis, Zhepei Wang, Paris Smaragdis

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

Abstract

In this paper, we present an efficient neural network for end-to-end general purpose audio source separation. Specifically, the backbone structure of this convolutional network is the SUccessive DOwnsampling and Resampling of Multi-Resolution Features (SuDoRM-RF) as well as their aggregation which is performed through simple one-dimensional convolutions. In this way, we are able to obtain high quality audio source separation with limited number of floating point operations, memory requirements, number of parameters and latency. Our experiments on both speech and environmental sound separation datasets show that SuDoRM - RF performs comparably and even surpasses various state-of-the-art approaches with significantly higher computational resource requirements.

Original language	English (US)
Title of host publication	Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020
Publisher	IEEE Computer Society
ISBN (Electronic)	9781728166629
DOIs	https://doi.org/10.1109/MLSP49062.2020.9231900
State	Published - Sep 2020
Externally published	Yes
Event	30th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2020 - Virtual, Espoo, Finland Duration: Sep 21 2020 → Sep 24 2020

Publication series

Name	IEEE International Workshop on Machine Learning for Signal Processing, MLSP
Volume	2020-September
ISSN (Print)	2161-0363
ISSN (Electronic)	2161-0371

Conference

Conference	30th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2020
Country/Territory	Finland
City	Virtual, Espoo
Period	9/21/20 → 9/24/20

Keywords

Audio source separation
Deep learning
Low-cost neural networks

ASJC Scopus subject areas

Human-Computer Interaction
Signal Processing

Online availability

10.1109/MLSP49062.2020.9231900

Library availability

Discover UIUC Full Text

Cite this

Tzinis, E., Wang, Z., & Smaragdis, P. (2020). Sudo RM -RF: Efficient networks for universal audio source separation. In Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020 Article 9231900 (IEEE International Workshop on Machine Learning for Signal Processing, MLSP; Vol. 2020-September). IEEE Computer Society. https://doi.org/10.1109/MLSP49062.2020.9231900

Sudo RM -RF: Efficient networks for universal audio source separation. / Tzinis, Efthymios; Wang, Zhepei; Smaragdis, Paris.
Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020. IEEE Computer Society, 2020. 9231900 (IEEE International Workshop on Machine Learning for Signal Processing, MLSP; Vol. 2020-September).

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

Tzinis, E, Wang, Z & Smaragdis, P 2020, Sudo RM -RF: Efficient networks for universal audio source separation. in Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020., 9231900, IEEE International Workshop on Machine Learning for Signal Processing, MLSP, vol. 2020-September, IEEE Computer Society, 30th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2020, Virtual, Espoo, Finland, 9/21/20. https://doi.org/10.1109/MLSP49062.2020.9231900

@inproceedings{b0a5dfa18e1145bfb83c7402ac43a36d,

title = "Sudo RM -RF: Efficient networks for universal audio source separation",

abstract = "In this paper, we present an efficient neural network for end-to-end general purpose audio source separation. Specifically, the backbone structure of this convolutional network is the SUccessive DOwnsampling and Resampling of Multi-Resolution Features (SuDoRM-RF) as well as their aggregation which is performed through simple one-dimensional convolutions. In this way, we are able to obtain high quality audio source separation with limited number of floating point operations, memory requirements, number of parameters and latency. Our experiments on both speech and environmental sound separation datasets show that SuDoRM - RF performs comparably and even surpasses various state-of-the-art approaches with significantly higher computational resource requirements.",

keywords = "Audio source separation, Deep learning, Low-cost neural networks",

author = "Efthymios Tzinis and Zhepei Wang and Paris Smaragdis",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 30th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2020 ; Conference date: 21-09-2020 Through 24-09-2020",

year = "2020",

month = sep,

doi = "10.1109/MLSP49062.2020.9231900",

language = "English (US)",

series = "IEEE International Workshop on Machine Learning for Signal Processing, MLSP",

publisher = "IEEE Computer Society",

booktitle = "Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020",

}

TY - GEN

T1 - Sudo RM -RF

T2 - 30th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2020

AU - Tzinis, Efthymios

AU - Wang, Zhepei

AU - Smaragdis, Paris

PY - 2020/9

Y1 - 2020/9

N2 - In this paper, we present an efficient neural network for end-to-end general purpose audio source separation. Specifically, the backbone structure of this convolutional network is the SUccessive DOwnsampling and Resampling of Multi-Resolution Features (SuDoRM-RF) as well as their aggregation which is performed through simple one-dimensional convolutions. In this way, we are able to obtain high quality audio source separation with limited number of floating point operations, memory requirements, number of parameters and latency. Our experiments on both speech and environmental sound separation datasets show that SuDoRM - RF performs comparably and even surpasses various state-of-the-art approaches with significantly higher computational resource requirements.

AB - In this paper, we present an efficient neural network for end-to-end general purpose audio source separation. Specifically, the backbone structure of this convolutional network is the SUccessive DOwnsampling and Resampling of Multi-Resolution Features (SuDoRM-RF) as well as their aggregation which is performed through simple one-dimensional convolutions. In this way, we are able to obtain high quality audio source separation with limited number of floating point operations, memory requirements, number of parameters and latency. Our experiments on both speech and environmental sound separation datasets show that SuDoRM - RF performs comparably and even surpasses various state-of-the-art approaches with significantly higher computational resource requirements.

KW - Audio source separation

KW - Deep learning

KW - Low-cost neural networks

UR - http://www.scopus.com/inward/record.url?scp=85096506405&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85096506405&partnerID=8YFLogxK

U2 - 10.1109/MLSP49062.2020.9231900

DO - 10.1109/MLSP49062.2020.9231900

M3 - Conference contribution

AN - SCOPUS:85096506405

T3 - IEEE International Workshop on Machine Learning for Signal Processing, MLSP

BT - Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020

PB - IEEE Computer Society

Y2 - 21 September 2020 through 24 September 2020

ER -

Sudo RM -RF: Efficient networks for universal audio source separation

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this