Time-Frequency Networks for Audio Super-Resolution

Teck Yian Lim, Raymond A. Yeh, Yijia Xu, Minh N Do, Mark Allan Hasegawa-Johnson

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Audio super-resolution (a.k.a. bandwidth extension) is the challenging task of increasing the temporal resolution of audio signals. Recent deep networks approaches achieved promising results by modeling the task as a regression problem in either time or frequency domain. In this paper, we introduced Time-Frequency Network (TFNet), a deep network that utilizes supervision in both the time and frequency domain. We proposed a novel model architecture which allows the two domains to be jointly optimized. Results demonstrate that our method outperforms the state-of-the-art both quantitatively and qualitatively.

Original languageEnglish (US)
Title of host publication2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages646-650
Number of pages5
Volume2018-April
ISBN (Print)9781538646588
DOIs
StatePublished - Sep 10 2018
Event2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Calgary, Canada
Duration: Apr 15 2018Apr 20 2018

Other

Other2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018
CountryCanada
CityCalgary
Period4/15/184/20/18

Fingerprint

Bandwidth

Keywords

  • Audio super-resolution
  • Bandwidth extension
  • Deep learning

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Lim, T. Y., Yeh, R. A., Xu, Y., Do, M. N., & Hasegawa-Johnson, M. A. (2018). Time-Frequency Networks for Audio Super-Resolution. In 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings (Vol. 2018-April, pp. 646-650). [8462049] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2018.8462049

Time-Frequency Networks for Audio Super-Resolution. / Lim, Teck Yian; Yeh, Raymond A.; Xu, Yijia; Do, Minh N; Hasegawa-Johnson, Mark Allan.

2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Vol. 2018-April Institute of Electrical and Electronics Engineers Inc., 2018. p. 646-650 8462049.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lim, TY, Yeh, RA, Xu, Y, Do, MN & Hasegawa-Johnson, MA 2018, Time-Frequency Networks for Audio Super-Resolution. in 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. vol. 2018-April, 8462049, Institute of Electrical and Electronics Engineers Inc., pp. 646-650, 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018, Calgary, Canada, 4/15/18. https://doi.org/10.1109/ICASSP.2018.8462049
Lim TY, Yeh RA, Xu Y, Do MN, Hasegawa-Johnson MA. Time-Frequency Networks for Audio Super-Resolution. In 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Vol. 2018-April. Institute of Electrical and Electronics Engineers Inc. 2018. p. 646-650. 8462049 https://doi.org/10.1109/ICASSP.2018.8462049
Lim, Teck Yian ; Yeh, Raymond A. ; Xu, Yijia ; Do, Minh N ; Hasegawa-Johnson, Mark Allan. / Time-Frequency Networks for Audio Super-Resolution. 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Vol. 2018-April Institute of Electrical and Electronics Engineers Inc., 2018. pp. 646-650
@inproceedings{4e7d2b404aca4981bfc038f0aa588e92,
title = "Time-Frequency Networks for Audio Super-Resolution",
abstract = "Audio super-resolution (a.k.a. bandwidth extension) is the challenging task of increasing the temporal resolution of audio signals. Recent deep networks approaches achieved promising results by modeling the task as a regression problem in either time or frequency domain. In this paper, we introduced Time-Frequency Network (TFNet), a deep network that utilizes supervision in both the time and frequency domain. We proposed a novel model architecture which allows the two domains to be jointly optimized. Results demonstrate that our method outperforms the state-of-the-art both quantitatively and qualitatively.",
keywords = "Audio super-resolution, Bandwidth extension, Deep learning",
author = "Lim, {Teck Yian} and Yeh, {Raymond A.} and Yijia Xu and Do, {Minh N} and Hasegawa-Johnson, {Mark Allan}",
year = "2018",
month = "9",
day = "10",
doi = "10.1109/ICASSP.2018.8462049",
language = "English (US)",
isbn = "9781538646588",
volume = "2018-April",
pages = "646--650",
booktitle = "2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - Time-Frequency Networks for Audio Super-Resolution

AU - Lim, Teck Yian

AU - Yeh, Raymond A.

AU - Xu, Yijia

AU - Do, Minh N

AU - Hasegawa-Johnson, Mark Allan

PY - 2018/9/10

Y1 - 2018/9/10

N2 - Audio super-resolution (a.k.a. bandwidth extension) is the challenging task of increasing the temporal resolution of audio signals. Recent deep networks approaches achieved promising results by modeling the task as a regression problem in either time or frequency domain. In this paper, we introduced Time-Frequency Network (TFNet), a deep network that utilizes supervision in both the time and frequency domain. We proposed a novel model architecture which allows the two domains to be jointly optimized. Results demonstrate that our method outperforms the state-of-the-art both quantitatively and qualitatively.

AB - Audio super-resolution (a.k.a. bandwidth extension) is the challenging task of increasing the temporal resolution of audio signals. Recent deep networks approaches achieved promising results by modeling the task as a regression problem in either time or frequency domain. In this paper, we introduced Time-Frequency Network (TFNet), a deep network that utilizes supervision in both the time and frequency domain. We proposed a novel model architecture which allows the two domains to be jointly optimized. Results demonstrate that our method outperforms the state-of-the-art both quantitatively and qualitatively.

KW - Audio super-resolution

KW - Bandwidth extension

KW - Deep learning

UR - http://www.scopus.com/inward/record.url?scp=85054250702&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054250702&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2018.8462049

DO - 10.1109/ICASSP.2018.8462049

M3 - Conference contribution

AN - SCOPUS:85054250702

SN - 9781538646588

VL - 2018-April

SP - 646

EP - 650

BT - 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -