A Novel Deep Learning Approach for Classification of Bird Sound Using Mel Frequency Cepstral Coefficients

Monitoring animal populations is one important matter to better understand changes in theirpopulation, behavior, and biodiversity. Bird sounds are the main tool to classify bird species acoustically. Thesounds of birds are an indicator for ecologists as it responds to changes in their environment. The recognitionamong a variety of bird species to get important features is computationally expensive. With the unbalancedclasses and scarcity of training data, the performance accuracy is degrading. This paper aims to classify species ofbirds using lightweight convolutional neural networks (LWCNNs) basis on using a spectrogram image of Brazilianbird sounds as a dataset. For extracting spectrogram images, Mel Frequency Cepstral Coefficient (MFCC)algorithm is used. To prove the high performance of the classifier, ten species of birds with 10,000 spectrogramimages are provided to the classifier. Our LWCNN model achieved a training and testing accuracy of 99.68 % and92.80 % respectively in 10.54 min with 5 epochs.