Speech Dereverberation Based on Convex Optimization Algorithms for Group Sparse Linear Prediction

Daniele Giacobello, Tobias Lindstrøm Jensen

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

2 Citations (Scopus)
126 Downloads (Pure)

Abstract

In this paper, we consider methods for improving far-field speech
recognition using dereverberation based on sparse multi-channel linear
prediction. In particular, we extend successful methods based on nonconvex iteratively reweighted least squares, that look for a sparse
desired speech signal in the short-term Fourier transform domain, by proposing sparsity promoting convex functions. Furthermore, we show how to improve performance by applying regularization into both the reweighted least squares and convex methods. We evaluate
the methods using large scale simulations by mimicking the application scenarios of interest. The experiments show that the proposed
convex formulations and regularization offer improvements over existing methods with added robustness and flexibility in fairly different
acoustic scenarios.
Original languageEnglish
Title of host publicationIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018
Number of pages5
PublisherIEEE
Publication date2018
Pages446-450
ISBN (Print)978-1-5386-4659-5
ISBN (Electronic)978-1-5386-4658-8
DOIs
Publication statusPublished - 2018
Event2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) - Calgary, Canada
Duration: 15 Apr 201820 Apr 2018
https://2018.ieeeicassp.org/

Conference

Conference2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
CountryCanada
CityCalgary
Period15/04/201820/04/2018
Internet address
SeriesI E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings
ISSN1520-6149

Cite this

Giacobello, D., & Jensen, T. L. (2018). Speech Dereverberation Based on Convex Optimization Algorithms for Group Sparse Linear Prediction. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018 (pp. 446-450). IEEE. I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings https://doi.org/10.1109/ICASSP.2018.8462560
Giacobello, Daniele ; Jensen, Tobias Lindstrøm. / Speech Dereverberation Based on Convex Optimization Algorithms for Group Sparse Linear Prediction. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018. IEEE, 2018. pp. 446-450 (I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings).
@inproceedings{d787ac0c84114f21bfaed63a76f8cc86,
title = "Speech Dereverberation Based on Convex Optimization Algorithms for Group Sparse Linear Prediction",
abstract = "In this paper, we consider methods for improving far-field speechrecognition using dereverberation based on sparse multi-channel linearprediction. In particular, we extend successful methods based on nonconvex iteratively reweighted least squares, that look for a sparsedesired speech signal in the short-term Fourier transform domain, by proposing sparsity promoting convex functions. Furthermore, we show how to improve performance by applying regularization into both the reweighted least squares and convex methods. We evaluatethe methods using large scale simulations by mimicking the application scenarios of interest. The experiments show that the proposedconvex formulations and regularization offer improvements over existing methods with added robustness and flexibility in fairly differentacoustic scenarios.",
author = "Daniele Giacobello and Jensen, {Tobias Lindstr{\o}m}",
year = "2018",
doi = "10.1109/ICASSP.2018.8462560",
language = "English",
isbn = "978-1-5386-4659-5",
pages = "446--450",
booktitle = "IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018",
publisher = "IEEE",
address = "United States",

}

Giacobello, D & Jensen, TL 2018, Speech Dereverberation Based on Convex Optimization Algorithms for Group Sparse Linear Prediction. in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018. IEEE, I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings, pp. 446-450, 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Calgary, Canada, 15/04/2018. https://doi.org/10.1109/ICASSP.2018.8462560

Speech Dereverberation Based on Convex Optimization Algorithms for Group Sparse Linear Prediction. / Giacobello, Daniele; Jensen, Tobias Lindstrøm.

IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018. IEEE, 2018. p. 446-450 (I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings).

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Speech Dereverberation Based on Convex Optimization Algorithms for Group Sparse Linear Prediction

AU - Giacobello, Daniele

AU - Jensen, Tobias Lindstrøm

PY - 2018

Y1 - 2018

N2 - In this paper, we consider methods for improving far-field speechrecognition using dereverberation based on sparse multi-channel linearprediction. In particular, we extend successful methods based on nonconvex iteratively reweighted least squares, that look for a sparsedesired speech signal in the short-term Fourier transform domain, by proposing sparsity promoting convex functions. Furthermore, we show how to improve performance by applying regularization into both the reweighted least squares and convex methods. We evaluatethe methods using large scale simulations by mimicking the application scenarios of interest. The experiments show that the proposedconvex formulations and regularization offer improvements over existing methods with added robustness and flexibility in fairly differentacoustic scenarios.

AB - In this paper, we consider methods for improving far-field speechrecognition using dereverberation based on sparse multi-channel linearprediction. In particular, we extend successful methods based on nonconvex iteratively reweighted least squares, that look for a sparsedesired speech signal in the short-term Fourier transform domain, by proposing sparsity promoting convex functions. Furthermore, we show how to improve performance by applying regularization into both the reweighted least squares and convex methods. We evaluatethe methods using large scale simulations by mimicking the application scenarios of interest. The experiments show that the proposedconvex formulations and regularization offer improvements over existing methods with added robustness and flexibility in fairly differentacoustic scenarios.

U2 - 10.1109/ICASSP.2018.8462560

DO - 10.1109/ICASSP.2018.8462560

M3 - Article in proceeding

SN - 978-1-5386-4659-5

SP - 446

EP - 450

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

PB - IEEE

ER -

Giacobello D, Jensen TL. Speech Dereverberation Based on Convex Optimization Algorithms for Group Sparse Linear Prediction. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018. IEEE. 2018. p. 446-450. (I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings). https://doi.org/10.1109/ICASSP.2018.8462560