Fast recognition of bird sounds using extreme learning machines

Kun Qian; Jian Guo; Ken Ishida; Satoshi Matsuoka

doi:10.1002/tee.22378

Fast recognition of bird sounds using extreme learning machines

Kun Qian^*, Jian Guo, Ken Ishida, Satoshi Matsuoka

^*Corresponding author for this work

Research output: Contribution to journal › Letter › peer-review

1 Citation (Scopus)

Abstract

Recognition of bird species by their sounds can bring considerable significance to both ecologists and ornithologists for measuring the biodiversity in the reserves, and studying climate changes. In this letter, we propose an efficient method based on an extreme learning machine (ELM) to classify bird sounds of 86 species of birds in very limited training and testing time. Experimental results prove that, the proposed ELM method can achieve the best recognition performance (81.1 %, unweighted average recall) compared with K-nearest neighbours (K-NN), support vector machines (SVM), neural networks (NN), and deep neural networks (DNN) pre-trained by an autoencoder. In addition, ELM requires the least total time for training and testing (2.047 ± 0.034 s).

Original language	English
Pages (from-to)	294-296
Number of pages	3
Journal	IEEJ Transactions on Electrical and Electronic Engineering
Volume	12
Issue number	2
DOIs	https://doi.org/10.1002/tee.22378
Publication status	Published - 1 Mar 2017
Externally published	Yes

Keywords

bio-acoustics
bird sounds
ecology
extreme learning machines
openSMILE

UN SDGs

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

Access to Document

10.1002/tee.22378

Cite this

@article{8f16710de4b14f60bd0a4e717bc7199e,

title = "Fast recognition of bird sounds using extreme learning machines",

abstract = "Recognition of bird species by their sounds can bring considerable significance to both ecologists and ornithologists for measuring the biodiversity in the reserves, and studying climate changes. In this letter, we propose an efficient method based on an extreme learning machine (ELM) to classify bird sounds of 86 species of birds in very limited training and testing time. Experimental results prove that, the proposed ELM method can achieve the best recognition performance (81.1 %, unweighted average recall) compared with K-nearest neighbours (K-NN), support vector machines (SVM), neural networks (NN), and deep neural networks (DNN) pre-trained by an autoencoder. In addition, ELM requires the least total time for training and testing (2.047 ± 0.034 s).",

keywords = "bio-acoustics, bird sounds, ecology, extreme learning machines, openSMILE",

author = "Kun Qian and Jian Guo and Ken Ishida and Satoshi Matsuoka",

note = "Publisher Copyright: {\textcopyright} 2016 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.",

year = "2017",

month = mar,

day = "1",

doi = "10.1002/tee.22378",

language = "English",

volume = "12",

pages = "294--296",

journal = "IEEJ Transactions on Electrical and Electronic Engineering",

issn = "1931-4973",

publisher = "John Wiley & Sons Inc.",

number = "2",

}

TY - JOUR

T1 - Fast recognition of bird sounds using extreme learning machines

AU - Qian, Kun

AU - Guo, Jian

AU - Ishida, Ken

AU - Matsuoka, Satoshi

PY - 2017/3/1

Y1 - 2017/3/1

N2 - Recognition of bird species by their sounds can bring considerable significance to both ecologists and ornithologists for measuring the biodiversity in the reserves, and studying climate changes. In this letter, we propose an efficient method based on an extreme learning machine (ELM) to classify bird sounds of 86 species of birds in very limited training and testing time. Experimental results prove that, the proposed ELM method can achieve the best recognition performance (81.1 %, unweighted average recall) compared with K-nearest neighbours (K-NN), support vector machines (SVM), neural networks (NN), and deep neural networks (DNN) pre-trained by an autoencoder. In addition, ELM requires the least total time for training and testing (2.047 ± 0.034 s).

AB - Recognition of bird species by their sounds can bring considerable significance to both ecologists and ornithologists for measuring the biodiversity in the reserves, and studying climate changes. In this letter, we propose an efficient method based on an extreme learning machine (ELM) to classify bird sounds of 86 species of birds in very limited training and testing time. Experimental results prove that, the proposed ELM method can achieve the best recognition performance (81.1 %, unweighted average recall) compared with K-nearest neighbours (K-NN), support vector machines (SVM), neural networks (NN), and deep neural networks (DNN) pre-trained by an autoencoder. In addition, ELM requires the least total time for training and testing (2.047 ± 0.034 s).

KW - bio-acoustics

KW - bird sounds

KW - ecology

KW - extreme learning machines

KW - openSMILE

UR - http://www.scopus.com/inward/record.url?scp=85008262691&partnerID=8YFLogxK

U2 - 10.1002/tee.22378

DO - 10.1002/tee.22378

M3 - Letter

AN - SCOPUS:85008262691

SN - 1931-4973

VL - 12

SP - 294

EP - 296

JO - IEEJ Transactions on Electrical and Electronic Engineering

JF - IEEJ Transactions on Electrical and Electronic Engineering

IS - 2

ER -

Fast recognition of bird sounds using extreme learning machines

Abstract

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this