Title :
Multi-label bird classification using an ensemble classifier with simple features
Author :
Yi Ren Leng ; Huy Dat Tran
Author_Institution :
Inst. for Infocomm Res., A*STAR, Singapore, Singapore
Abstract :
For the Multi-label Bird Species Classification task in NIPS 2013, we design an ensemble classifier using simple audio features extracted from the given audio clips. The main difference of our approach is that we do not segment the audio or use local features and still achieve comparable results to the top teams of the challenge utilizing complex engineered features.
Keywords :
audio signal processing; feature extraction; signal classification; audio clips; audio feature extraction; ensemble classifier; multilabel bird species classification task; Birds; Feature extraction; Mel frequency cepstral coefficient; Radio frequency; Training; Vegetation;
Conference_Titel :
Asia-Pacific Signal and Information Processing Association, 2014 Annual Summit and Conference (APSIPA)
Conference_Location :
Siem Reap
DOI :
10.1109/APSIPA.2014.7041649