Extraction of acoustic features of calls of five Platypleura species using the field recordings

Abstract

Five Platypleura species are distributed in the south of Amami Island in Japan. At particular, P. albivannata, and P. yayeyamana are distributed at the same area in Ishigaki island. The survey of P. albivannata, which is endangered species, is important problems for conservation. Although it is necessary to discriminate between these species, it is difficult to discriminate between them from the morphology, that is, visual information. It is necessary to discriminate with these species by using another cue. In this presentation, we proposed the analysis methods to extract acoustic features dependent on Platypleura species to the south of Amami Island in Japan. We primarily aimed to examine whether these classifications correspond to the resemblance in the time and frequency domain of calling songs. Furthermore, we tried to extract characteristics that can be used for species identification in monitoring endangered species, P. albivannata. We found that maximum peak frequencies are feasible for species identification, especially for the detection of P. albivannata in the wild.