Automatic recognition of birdsongs using mel-frequency cepstral coefficients and vector quantization

In this paper we propose a method to automatically identify birds from the sounds they generate. First, each syllable corresponding to a piece of vocalization is segmented. For each syllable, the averaged MFCCs (AMFCC) over all frames in a syllable are calculated as the vocalization features. Vector quantization is employed to design the codebook consisting of several representative feature vectors to well model the variant characteristics of syllables segmented from the sound of identical bird species. Finally, linear discriminant analysis (LDA) is exploited to increase the classification accuracy at a lower dimensional feature vector space. In our experiments, the average classification accuracy is 87% for the recognition of 420 bird species.