Greek Folk Music Denoising Under a Symmetric α-Stable Noise Assumption

Nikoletta  Bassiou; Constantine  Kotropoulos; Ioannis Pitas

doi:10.1109/QSHINE.2014.6928654

Greek Folk Music Denoising Under a Symmetric α-Stable Noise Assumption

Nikoletta Bassiou, Constantine Kotropoulos, Ioannis Pitas

Department of Electrical & Electronic Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

2 Citations (Scopus)

275 Downloads (Pure)

Abstract

The noise in musical audio recordings is assumed to obey an α-stable distribution. A sparse linear regression framework with structured priors is elaborated. Markov Chain Monte Carlo is used to infer the clean music signal model and the α-stable noise distribution parameters. The musical audio recordings are processed both as a whole and in segments by using a sine-bell window for analysis and overlap-and-add reconstruction. Experiments on noisy Greek folk music excerpts demonstrate better denoising under the α-stable noise assumption than the Gaussian white noise one, and when processing is performed in segments rather than in full recordings.

Original language	English
Title of host publication	2014 10th International Conference on Heterogeneous Networking for Quality, Reliability, Security and Robustness (QShine 2014)
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	18-23
Number of pages	6
ISBN (Print)	9781509000111
DOIs	https://doi.org/10.1109/QSHINE.2014.6928654
Publication status	Published - 18 Aug 2014
Event	10th International Conference Heterogeneous Networking for Quality, Reliability, Security and Robustness (QSHINE 2014 - Rhodes, Greece Duration: 18 Aug 2014 → 20 Aug 2014

Conference

Conference	10th International Conference Heterogeneous Networking for Quality, Reliability, Security and Robustness (QSHINE 2014
Country/Territory	Greece
City	Rhodes
Period	18/08/14 → 20/08/14

Access to Document

10.1109/QSHINE.2014.6928654

Ionnas Pitas - Greek Folk Music Denoising Under a Symmetric a-Stable Noise Assumption - 2014
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IEEE at http://dx.doi.org/10.1109/QSHINE.2014.6928654. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 2.28 MB

Handle.net

Persistent link

Cite this

@inproceedings{93bb5c603c3c4c6a9f4127bb25b37957,

title = "Greek Folk Music Denoising Under a Symmetric α-Stable Noise Assumption",

abstract = "The noise in musical audio recordings is assumed to obey an α-stable distribution. A sparse linear regression framework with structured priors is elaborated. Markov Chain Monte Carlo is used to infer the clean music signal model and the α-stable noise distribution parameters. The musical audio recordings are processed both as a whole and in segments by using a sine-bell window for analysis and overlap-and-add reconstruction. Experiments on noisy Greek folk music excerpts demonstrate better denoising under the α-stable noise assumption than the Gaussian white noise one, and when processing is performed in segments rather than in full recordings.",

author = "Nikoletta Bassiou and Constantine Kotropoulos and Ioannis Pitas",

year = "2014",

month = aug,

day = "18",

doi = "10.1109/QSHINE.2014.6928654",

language = "English",

isbn = "9781509000111",

pages = "18--23",

booktitle = "2014 10th International Conference on Heterogeneous Networking for Quality, Reliability, Security and Robustness (QShine 2014)",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

note = "10th International Conference Heterogeneous Networking for Quality, Reliability, Security and Robustness (QSHINE 2014 ; Conference date: 18-08-2014 Through 20-08-2014",

}

Bassiou, N, Kotropoulos, C & Pitas, I 2014, Greek Folk Music Denoising Under a Symmetric α-Stable Noise Assumption. in 2014 10th International Conference on Heterogeneous Networking for Quality, Reliability, Security and Robustness (QShine 2014). Institute of Electrical and Electronics Engineers (IEEE), pp. 18-23, 10th International Conference Heterogeneous Networking for Quality, Reliability, Security and Robustness (QSHINE 2014, Rhodes, Greece, 18/08/14. https://doi.org/10.1109/QSHINE.2014.6928654

Greek Folk Music Denoising Under a Symmetric α-Stable Noise Assumption. / Bassiou, Nikoletta ; Kotropoulos, Constantine ; Pitas, Ioannis.
2014 10th International Conference on Heterogeneous Networking for Quality, Reliability, Security and Robustness (QShine 2014). Institute of Electrical and Electronics Engineers (IEEE), 2014. p. 18-23.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Greek Folk Music Denoising Under a Symmetric α-Stable Noise Assumption

AU - Bassiou, Nikoletta

AU - Kotropoulos, Constantine

AU - Pitas, Ioannis

PY - 2014/8/18

Y1 - 2014/8/18

N2 - The noise in musical audio recordings is assumed to obey an α-stable distribution. A sparse linear regression framework with structured priors is elaborated. Markov Chain Monte Carlo is used to infer the clean music signal model and the α-stable noise distribution parameters. The musical audio recordings are processed both as a whole and in segments by using a sine-bell window for analysis and overlap-and-add reconstruction. Experiments on noisy Greek folk music excerpts demonstrate better denoising under the α-stable noise assumption than the Gaussian white noise one, and when processing is performed in segments rather than in full recordings.

AB - The noise in musical audio recordings is assumed to obey an α-stable distribution. A sparse linear regression framework with structured priors is elaborated. Markov Chain Monte Carlo is used to infer the clean music signal model and the α-stable noise distribution parameters. The musical audio recordings are processed both as a whole and in segments by using a sine-bell window for analysis and overlap-and-add reconstruction. Experiments on noisy Greek folk music excerpts demonstrate better denoising under the α-stable noise assumption than the Gaussian white noise one, and when processing is performed in segments rather than in full recordings.

U2 - 10.1109/QSHINE.2014.6928654

DO - 10.1109/QSHINE.2014.6928654

M3 - Conference Contribution (Conference Proceeding)

SN - 9781509000111

SP - 18

EP - 23

BT - 2014 10th International Conference on Heterogeneous Networking for Quality, Reliability, Security and Robustness (QShine 2014)

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 10th International Conference Heterogeneous Networking for Quality, Reliability, Security and Robustness (QSHINE 2014

Y2 - 18 August 2014 through 20 August 2014

ER -

Greek Folk Music Denoising Under a Symmetric α-Stable Noise Assumption

Abstract

Conference

Access to Document

Handle.net

Fingerprint

Cite this