Recognizing bird species in audio recordings using deep convolutional neural networks

Karol Jerzy Piczak

Abstract

This paper summarizes a method for purely audio-based bird species recognition through the application of convolutional neural networks. The approach is evaluated in the context of the LifeCLEF 2016 bird identi�cation task - an open challenge conducted on a dataset containing 34 128 audio recordings representing 999 bird species from South America. Three di�erent network architectures and a simple ensemble model are considered for this task, with the ensemble submission achieving a mean average precision of 41.2% (o�cial score) and 52.9% for foreground species.
Author Karol Jerzy Piczak II
Karol Jerzy Piczak,,
- The Institute of Computer Science
Pages534-543
Publication size in sheets0.5
Book Balog Krisztian, Cappellato Linda, Ferro Nicola, Macdonald Craig (eds.): Working Notes of CLEF 2016 - Conference and Labs of the Evaluation forum, CEUR Workshop Proceedings, vol. 1609, 2016, CEUR-WS, 1259 p.
Keywords in Englishbird species identi�cation, convolutional neural networks, audio classi�cation, BirdCLEF 2016
URL http://ceur-ws.org/Vol-1609/16090534.pdf
Languageen angielski
File
16090534.pdf / 1.43 MB / 16090534.pdf 1.43 MB
Score (nominal)0
ScoreMinisterial score = 0.0, 27-03-2017, BookChapterNotSeriesMainLanguages
Citation count*0
Cite
Share Share



* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Back