Feature Extraction of Binaural Recordings for Acoustic Scene Classification

Sławomir Zieliński; Hyunkook Lee

Feature Extraction of Binaural Recordings for Acoustic Scene Classification

Sławomir Zieliński, Hyunkook Lee

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

Abstract

Binaural technology becomes increasingly popular in the multimedia systems. This paper identifies a set of features of binaural recordings suitable for the automatic classification of the four basic spatial audio scenes representing the most typical patterns of audio content distribution around a listener. Moreover, it compares the five artificial-intelligence-based methods applied to the classification of binaural recordings. The results show that both the spatial and the spectro-temporal features are essential to accurate classification of binaurally rendered acoustic scenes. The spectro-temporal features appear to have a stronger influence on the classification results than the spatial metrics. According to the obtained results, the method based on the support vector machine, exploiting the features identified in the study, yields the classification accuracy approaching 84%.

Original language	English
Title of host publication	IEEE Federated Conference on Computer Science and Information Systems, FedCSIS 2018
Editors	Maria Ganzha, Leszek Maciaszek, Marcin Paprzycki
Publisher	IEEE
Pages	585-588
Number of pages	4
ISBN (Electronic)	9788394941956
ISBN (Print)	9781538623718
Publication status	Published - 29 Oct 2018
Event	IEEE Federated Conference on Computer Science and Information Systems - Poznan, Poland Duration: 9 Sep 2018 → 12 Sep 2018 https://fedcsis.org/2018/ (Link to Conference Website)

Conference

Conference	IEEE Federated Conference on Computer Science and Information Systems
Abbreviated title	FedCSIS
Country/Territory	Poland
City	Poznan
Period	9/09/18 → 12/09/18
Internet address	https://fedcsis.org/2018/ (Link to Conference Website)

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

IEEE_2018_SpatialSceneClassification_Zilensky_and_LeeAccepted author manuscript, 292 KBLicence: Unspecified

https://ieeexplore.ieee.org/document/8511268Licence: Unspecified

Cite this

@inproceedings{ad6bcdce03a74fbc9c764f6298f9f418,

title = "Feature Extraction of Binaural Recordings for Acoustic Scene Classification",

abstract = "Binaural technology becomes increasingly popular in the multimedia systems. This paper identifies a set of features of binaural recordings suitable for the automatic classification of the four basic spatial audio scenes representing the most typical patterns of audio content distribution around a listener. Moreover, it compares the five artificial-intelligence-based methods applied to the classification of binaural recordings. The results show that both the spatial and the spectro-temporal features are essential to accurate classification of binaurally rendered acoustic scenes. The spectro-temporal features appear to have a stronger influence on the classification results than the spatial metrics. According to the obtained results, the method based on the support vector machine, exploiting the features identified in the study, yields the classification accuracy approaching 84%.",

keywords = "Spatial scene, Classification, Machine learning, Artificial Intelligence, Binaural audio",

author = "S{\l}awomir Zieli{\'n}ski and Hyunkook Lee",

note = "{\textcopyright} {\textcopyright} 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.; IEEE Federated Conference on Computer Science and Information Systems<br/>, FedCSIS ; Conference date: 09-09-2018 Through 12-09-2018",

year = "2018",

month = oct,

day = "29",

language = "English",

isbn = "9781538623718",

pages = "585--588",

editor = "Maria Ganzha and Leszek Maciaszek and Marcin Paprzycki",

booktitle = "IEEE Federated Conference on Computer Science and Information Systems, FedCSIS 2018",

publisher = "IEEE",

address = "United States",

url = "https://fedcsis.org/2018/",

}

Zieliński, S & Lee, H 2018, Feature Extraction of Binaural Recordings for Acoustic Scene Classification. in M Ganzha, L Maciaszek & M Paprzycki (eds), IEEE Federated Conference on Computer Science and Information Systems, FedCSIS 2018. IEEE, pp. 585-588, IEEE Federated Conference on Computer Science and Information Systems
, Poznan, Poland, 9/09/18. <https://ieeexplore.ieee.org/document/8511268>

Feature Extraction of Binaural Recordings for Acoustic Scene Classification. / Zieliński, Sławomir; Lee, Hyunkook.

IEEE Federated Conference on Computer Science and Information Systems, FedCSIS 2018. ed. / Maria Ganzha; Leszek Maciaszek; Marcin Paprzycki. IEEE, 2018. p. 585-588.

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

TY - GEN

T1 - Feature Extraction of Binaural Recordings for Acoustic Scene Classification

AU - Zieliński, Sławomir

AU - Lee, Hyunkook

N1 - © © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2018/10/29

Y1 - 2018/10/29

N2 - Binaural technology becomes increasingly popular in the multimedia systems. This paper identifies a set of features of binaural recordings suitable for the automatic classification of the four basic spatial audio scenes representing the most typical patterns of audio content distribution around a listener. Moreover, it compares the five artificial-intelligence-based methods applied to the classification of binaural recordings. The results show that both the spatial and the spectro-temporal features are essential to accurate classification of binaurally rendered acoustic scenes. The spectro-temporal features appear to have a stronger influence on the classification results than the spatial metrics. According to the obtained results, the method based on the support vector machine, exploiting the features identified in the study, yields the classification accuracy approaching 84%.

AB - Binaural technology becomes increasingly popular in the multimedia systems. This paper identifies a set of features of binaural recordings suitable for the automatic classification of the four basic spatial audio scenes representing the most typical patterns of audio content distribution around a listener. Moreover, it compares the five artificial-intelligence-based methods applied to the classification of binaural recordings. The results show that both the spatial and the spectro-temporal features are essential to accurate classification of binaurally rendered acoustic scenes. The spectro-temporal features appear to have a stronger influence on the classification results than the spatial metrics. According to the obtained results, the method based on the support vector machine, exploiting the features identified in the study, yields the classification accuracy approaching 84%.

KW - Spatial scene

KW - Classification

KW - Machine learning

KW - Artificial Intelligence

KW - Binaural audio

M3 - Conference contribution

SN - 9781538623718

SP - 585

EP - 588

BT - IEEE Federated Conference on Computer Science and Information Systems, FedCSIS 2018

A2 - Ganzha, Maria

A2 - Maciaszek, Leszek

A2 - Paprzycki, Marcin

PB - IEEE

T2 - IEEE Federated Conference on Computer Science and Information Systems<br/>

Y2 - 9 September 2018 through 12 September 2018

ER -

Feature Extraction of Binaural Recordings for Acoustic Scene Classification

Abstract

Conference

UN SDGs

Access to Document

Fingerprint

Cite this