Modified DCTNet for audio signals classification

Yin Xian; Yunchen Pu; Zhe Gan; Liang Lu; Andrew Thompson

doi:10.1121/1.4970932

Modified DCTNet for audio signals classification

Yin Xian, Yunchen Pu, Zhe Gan, Liang Lu, Andrew Thompson

School of Continuing Education

Research output: Contribution to journal › Journal article › peer-review

Abstract

We introduced the use of modified DCTNet for audio signals feature extraction. The modified DCTNet is a development of DCTNet, with its center frequencies of filterbanks geometrically spaced. The modified DCTNet is adaptive to different acoustic scales, and it can better capture low frequency acoustic information that is sensitive to human audio perception. We use features extracted by the modified DCTNet and put them to classifiers. Experimental results on bird song classification, music genre classification, and artist identification show that the modified DCTNet and RNN improve classification rate, and achieve state of the art performance.

Original language	English
Article number	3405
Journal	Journal of the Acoustical Society of America
Volume	140
Issue number	4
DOIs	https://doi.org/10.1121/1.4970932
Publication status	Published - 1 Oct 2016

UN SDGs

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

Access to Document

10.1121/1.4970932

Cite this

@article{3aee07b5d4b5481cbf0e416c7308471c,

title = "Modified DCTNet for audio signals classification",

abstract = "We introduced the use of modified DCTNet for audio signals feature extraction. The modified DCTNet is a development of DCTNet, with its center frequencies of filterbanks geometrically spaced. The modified DCTNet is adaptive to different acoustic scales, and it can better capture low frequency acoustic information that is sensitive to human audio perception. We use features extracted by the modified DCTNet and put them to classifiers. Experimental results on bird song classification, music genre classification, and artist identification show that the modified DCTNet and RNN improve classification rate, and achieve state of the art performance.",

author = "Yin Xian and Yunchen Pu and Zhe Gan and Liang Lu and Andrew Thompson",

year = "2016",

month = oct,

day = "1",

doi = "10.1121/1.4970932",

language = "English",

volume = "140",

journal = "Journal of the Acoustical Society of America",

issn = "0001-4966",

publisher = "Acoustical Society of America",

number = "4",

}

TY - JOUR

T1 - Modified DCTNet for audio signals classification

AU - Xian, Yin

AU - Pu, Yunchen

AU - Gan, Zhe

AU - Lu, Liang

AU - Thompson, Andrew

PY - 2016/10/1

Y1 - 2016/10/1

N2 - We introduced the use of modified DCTNet for audio signals feature extraction. The modified DCTNet is a development of DCTNet, with its center frequencies of filterbanks geometrically spaced. The modified DCTNet is adaptive to different acoustic scales, and it can better capture low frequency acoustic information that is sensitive to human audio perception. We use features extracted by the modified DCTNet and put them to classifiers. Experimental results on bird song classification, music genre classification, and artist identification show that the modified DCTNet and RNN improve classification rate, and achieve state of the art performance.

AB - We introduced the use of modified DCTNet for audio signals feature extraction. The modified DCTNet is a development of DCTNet, with its center frequencies of filterbanks geometrically spaced. The modified DCTNet is adaptive to different acoustic scales, and it can better capture low frequency acoustic information that is sensitive to human audio perception. We use features extracted by the modified DCTNet and put them to classifiers. Experimental results on bird song classification, music genre classification, and artist identification show that the modified DCTNet and RNN improve classification rate, and achieve state of the art performance.

UR - https://pubs.aip.org/asa/jasa/article/140/4_Supplement/3405/713911/Modified-DCTNet-for-audio-signals-classification

U2 - 10.1121/1.4970932

DO - 10.1121/1.4970932

M3 - Journal article

SN - 0001-4966

VL - 140

JO - Journal of the Acoustical Society of America

JF - Journal of the Acoustical Society of America

IS - 4

M1 - 3405

ER -

Modified DCTNet for audio signals classification

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this