Visualization and categorization of ecological acoustic events based on discriminant features

Although sound classification in soundscape studies are generally performed by experts, the large growth of acoustic data presents a major challenge for performing such task. At same time, identification more discriminating features becomes crucial when analyzing soundscapes, and this occurs because natural anthropogenic sounds very complex, particularly Neotropical regions, where biodiversity level is high. In scenario, need research addressing discriminatory capability utmost importance to work towards automating these processes. study we present method identify most discriminant categorizing events soundscapes. Such key events. Our experimental findings validate our method, showing high certain extracted from data, reaching an accuracy 89.91% frogs, birds insects simultaneously. An extension experiments simulate binary reached 82.64%,100.0% 99.40% between combinations frogs-birds, frogs-insects birds-insects, respectively.