Environmental sound classification using hybrid SVM/KNN classifier and MPEG-7 audio low-level descriptor

Jia Ching Wang; Jhing Fa Wang; Kuok Wai He; Cheng Shu Hsu

Environmental sound classification using hybrid SVM/KNN classifier and MPEG-7 audio low-level descriptor

Jia Ching Wang, Jhing Fa Wang, Kuok Wai He, Cheng Shu Hsu

資訊工程學系

研究成果: 書貢獻/報告類型 › 會議論文篇章 › 同行評審

103 引文斯高帕斯（Scopus）

摘要

In this paper, we present a new environmental sound classification architecture. The proposed sound classifier is performed in frame level and fuses the support vector machine (SVM) and the k nearest neighbor rule (KNN). In feature selection, three MPEG-7 audio low-level descriptors, spectrum centroid, spectrum spread, and spectrum flatness are used as the sound features to exploit their ability in sound classification. Experiments carried out on a 12-class sound database can achieve an 85.1% accuracy rate. The performance comparison between the HMM sound classifier using audio spectrum projection features demonstrates the superiority of the proposed scheme.

原文	???core.languages.en_GB???
主出版物標題	International Joint Conference on Neural Networks 2006, IJCNN '06
頁面	1731-1735
頁數	5
出版狀態	已出版 - 2006
事件	International Joint Conference on Neural Networks 2006, IJCNN '06 - Vancouver, BC, Canada 持續時間: 16 7月 2006 → 21 7月 2006

出版系列

名字	IEEE International Conference on Neural Networks - Conference Proceedings
ISSN（列印）	1098-7576

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???	International Joint Conference on Neural Networks 2006, IJCNN '06
國家/地區	Canada
城市	Vancouver, BC
期間	16/07/06 → 21/07/06

引用此

@inproceedings{4a7fdd8b96ae4eeea66b3c5ad6c9e368,

title = "Environmental sound classification using hybrid SVM/KNN classifier and MPEG-7 audio low-level descriptor",

abstract = "In this paper, we present a new environmental sound classification architecture. The proposed sound classifier is performed in frame level and fuses the support vector machine (SVM) and the k nearest neighbor rule (KNN). In feature selection, three MPEG-7 audio low-level descriptors, spectrum centroid, spectrum spread, and spectrum flatness are used as the sound features to exploit their ability in sound classification. Experiments carried out on a 12-class sound database can achieve an 85.1% accuracy rate. The performance comparison between the HMM sound classifier using audio spectrum projection features demonstrates the superiority of the proposed scheme.",

author = "Wang, {Jia Ching} and Wang, {Jhing Fa} and He, {Kuok Wai} and Hsu, {Cheng Shu}",

year = "2006",

language = "???core.languages.en_GB???",

isbn = "0780394909",

series = "IEEE International Conference on Neural Networks - Conference Proceedings",

pages = "1731--1735",

booktitle = "International Joint Conference on Neural Networks 2006, IJCNN '06",

note = "International Joint Conference on Neural Networks 2006, IJCNN '06 ; Conference date: 16-07-2006 Through 21-07-2006",

}

Wang, JC, Wang, JF, He, KW & Hsu, CS 2006, Environmental sound classification using hybrid SVM/KNN classifier and MPEG-7 audio low-level descriptor. 於 International Joint Conference on Neural Networks 2006, IJCNN '06., 1716317, IEEE International Conference on Neural Networks - Conference Proceedings, 頁 1731-1735, International Joint Conference on Neural Networks 2006, IJCNN '06, Vancouver, BC, Canada, 16/07/06.

Environmental sound classification using hybrid SVM/KNN classifier and MPEG-7 audio low-level descriptor. / Wang, Jia Ching; Wang, Jhing Fa; He, Kuok Wai 等.
International Joint Conference on Neural Networks 2006, IJCNN '06. 2006. p. 1731-1735 1716317 (IEEE International Conference on Neural Networks - Conference Proceedings).

研究成果: 書貢獻/報告類型 › 會議論文篇章 › 同行評審

TY - GEN

T1 - Environmental sound classification using hybrid SVM/KNN classifier and MPEG-7 audio low-level descriptor

AU - Wang, Jia Ching

AU - Wang, Jhing Fa

AU - He, Kuok Wai

AU - Hsu, Cheng Shu

PY - 2006

Y1 - 2006

N2 - In this paper, we present a new environmental sound classification architecture. The proposed sound classifier is performed in frame level and fuses the support vector machine (SVM) and the k nearest neighbor rule (KNN). In feature selection, three MPEG-7 audio low-level descriptors, spectrum centroid, spectrum spread, and spectrum flatness are used as the sound features to exploit their ability in sound classification. Experiments carried out on a 12-class sound database can achieve an 85.1% accuracy rate. The performance comparison between the HMM sound classifier using audio spectrum projection features demonstrates the superiority of the proposed scheme.

AB - In this paper, we present a new environmental sound classification architecture. The proposed sound classifier is performed in frame level and fuses the support vector machine (SVM) and the k nearest neighbor rule (KNN). In feature selection, three MPEG-7 audio low-level descriptors, spectrum centroid, spectrum spread, and spectrum flatness are used as the sound features to exploit their ability in sound classification. Experiments carried out on a 12-class sound database can achieve an 85.1% accuracy rate. The performance comparison between the HMM sound classifier using audio spectrum projection features demonstrates the superiority of the proposed scheme.

UR - http://www.scopus.com/inward/record.url?scp=40649129039&partnerID=8YFLogxK

M3 - 會議論文篇章

AN - SCOPUS:40649129039

SN - 0780394909

SN - 9780780394902

T3 - IEEE International Conference on Neural Networks - Conference Proceedings

SP - 1731

EP - 1735

BT - International Joint Conference on Neural Networks 2006, IJCNN '06

T2 - International Joint Conference on Neural Networks 2006, IJCNN '06

Y2 - 16 July 2006 through 21 July 2006

ER -

Environmental sound classification using hybrid SVM/KNN classifier and MPEG-7 audio low-level descriptor

摘要

出版系列

???event.eventtypes.event.conference???

指紋

引用此