Sparse Modeling of The Early Part of Noisy Room Impulse Responses with Sparse Bayesian Learning

Maozhong Fu; Jesper Rindom Jensen; Yuhan Li; Mads Græsbøll Christensen

doi:10.1109/ICASSP43922.2022.9746069

Sparse Modeling of The Early Part of Noisy Room Impulse Responses with Sparse Bayesian Learning

Maozhong Fu, Jesper Rindom Jensen, Yuhan Li, Mads Græsbøll Christensen

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

5 Citations (Scopus)

Abstract

A model of a room impulse response (RIR) is useful for a wide range of applications. Typically, the early part of a RIR is sparse, and its sparse structure allows for accurate and simple modeling of the RIR. The existing p(0 < p ≤ 1)-norm-based methods suffer from the sensitivity to the user-selected regularization parameters or a high computational burden. In this work, we propose to reconstruct the sparse model for the early part of RIRs with sparse Bayesian learning (SBL). Under the framework of SBL, the proposed method can adaptively learn the optimal hyper-parameters from data at a low computational cost. Experiment results show that the proposed method has advantages in terms of noise robustness, reconstruction sparsity, and computational efficiency compared to the existing methods.

Original language	English
Title of host publication	2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
Number of pages	5
Publisher	IEEE
Publication date	2022
Pages	586-590
Article number	9746069
ISBN (Print)	978-1-6654-0541-6
ISBN (Electronic)	978-1-6654-0540-9
DOIs	https://doi.org/10.1109/ICASSP43922.2022.9746069
Publication status	Published - 2022
Event	47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Virtual, Online, Singapore Duration: 23 May 2022 → 27 May 2022

Conference

Conference	47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022
Country/Territory	Singapore
City	Virtual, Online
Period	23/05/2022 → 27/05/2022
Sponsor	Chinese and Oriental Languages Information Processing Society (COLPIS), Singapore Exhibition and Convention Bureau, The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen), The Institute of Electrical and Electronics Engineers Signal Processing Society

Series	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2022-May
ISSN	1520-6149

Bibliographical note

Publisher Copyright:
© 2022 IEEE

Keywords

Room impulse response
sparse Bayesian learning
sparse modeling

Access to Document

10.1109/ICASSP43922.2022.9746069

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

Fu, Maozhong ; Jensen, Jesper Rindom ; Li, Yuhan et al. / Sparse Modeling of The Early Part of Noisy Room Impulse Responses with Sparse Bayesian Learning. 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. IEEE, 2022. pp. 586-590 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Vol. 2022-May).

@inproceedings{676941b6bda243f7abf205d2f5e0dde7,

title = "Sparse Modeling of The Early Part of Noisy Room Impulse Responses with Sparse Bayesian Learning",

abstract = "A model of a room impulse response (RIR) is useful for a wide range of applications. Typically, the early part of a RIR is sparse, and its sparse structure allows for accurate and simple modeling of the RIR. The existing p(0 < p ≤ 1)-norm-based methods suffer from the sensitivity to the user-selected regularization parameters or a high computational burden. In this work, we propose to reconstruct the sparse model for the early part of RIRs with sparse Bayesian learning (SBL). Under the framework of SBL, the proposed method can adaptively learn the optimal hyper-parameters from data at a low computational cost. Experiment results show that the proposed method has advantages in terms of noise robustness, reconstruction sparsity, and computational efficiency compared to the existing methods.",

keywords = "Room impulse response, sparse Bayesian learning, sparse modeling",

author = "Maozhong Fu and Jensen, {Jesper Rindom} and Yuhan Li and Christensen, {Mads Gr{\ae}sb{\o}ll}",

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/ICASSP43922.2022.9746069",

language = "English",

isbn = "978-1-6654-0541-6",

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

publisher = "IEEE",

pages = "586--590",

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

address = "United States",

}

Fu, M, Jensen, JR, Li, Y & Christensen, MG 2022, Sparse Modeling of The Early Part of Noisy Room Impulse Responses with Sparse Bayesian Learning. in 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings., 9746069, IEEE, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2022-May, pp. 586-590, 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022, Virtual, Online, Singapore, 23/05/2022. https://doi.org/10.1109/ICASSP43922.2022.9746069

Sparse Modeling of The Early Part of Noisy Room Impulse Responses with Sparse Bayesian Learning. / Fu, Maozhong; Jensen, Jesper Rindom; Li, Yuhan et al.
2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. IEEE, 2022. p. 586-590 9746069 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Vol. 2022-May).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Sparse Modeling of The Early Part of Noisy Room Impulse Responses with Sparse Bayesian Learning

AU - Fu, Maozhong

AU - Jensen, Jesper Rindom

AU - Li, Yuhan

AU - Christensen, Mads Græsbøll

PY - 2022

Y1 - 2022

N2 - A model of a room impulse response (RIR) is useful for a wide range of applications. Typically, the early part of a RIR is sparse, and its sparse structure allows for accurate and simple modeling of the RIR. The existing p(0 < p ≤ 1)-norm-based methods suffer from the sensitivity to the user-selected regularization parameters or a high computational burden. In this work, we propose to reconstruct the sparse model for the early part of RIRs with sparse Bayesian learning (SBL). Under the framework of SBL, the proposed method can adaptively learn the optimal hyper-parameters from data at a low computational cost. Experiment results show that the proposed method has advantages in terms of noise robustness, reconstruction sparsity, and computational efficiency compared to the existing methods.

AB - A model of a room impulse response (RIR) is useful for a wide range of applications. Typically, the early part of a RIR is sparse, and its sparse structure allows for accurate and simple modeling of the RIR. The existing p(0 < p ≤ 1)-norm-based methods suffer from the sensitivity to the user-selected regularization parameters or a high computational burden. In this work, we propose to reconstruct the sparse model for the early part of RIRs with sparse Bayesian learning (SBL). Under the framework of SBL, the proposed method can adaptively learn the optimal hyper-parameters from data at a low computational cost. Experiment results show that the proposed method has advantages in terms of noise robustness, reconstruction sparsity, and computational efficiency compared to the existing methods.

KW - Room impulse response

KW - sparse Bayesian learning

KW - sparse modeling

UR - http://www.scopus.com/inward/record.url?scp=85131267185&partnerID=8YFLogxK

U2 - 10.1109/ICASSP43922.2022.9746069

DO - 10.1109/ICASSP43922.2022.9746069

M3 - Article in proceeding

AN - SCOPUS:85131267185

SN - 978-1-6654-0541-6

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

SP - 586

EP - 590

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

PB - IEEE

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

Y2 - 23 May 2022 through 27 May 2022

ER -

Fu M, Jensen JR, Li Y, Christensen MG. Sparse Modeling of The Early Part of Noisy Room Impulse Responses with Sparse Bayesian Learning. In 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. IEEE. 2022. p. 586-590. 9746069. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Vol. 2022-May). doi: 10.1109/ICASSP43922.2022.9746069

Sparse Modeling of The Early Part of Noisy Room Impulse Responses with Sparse Bayesian Learning

Abstract

Conference

Bibliographical note

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Cite this