Detection and classification of bottlenose dolphin’s (Tursiops truncatus, Montagu, 1821) vocalizations in the central-northern Adriatic Sea

This dataset is part of a collection and contains the results of dolphins’ vocalizations detection and classification. In details: <strong>whistles.txt</strong> includes useful information on 303 whistles visually detected from the whole recording, where: the first column reports the whistle starting time offset the second column reports the whistle end-time offset the third column contains the classification (W for single whistle, MW for multiple whistles) and the relative quality grade of the whistle. The whistle quality increases with increasing grade (from 1 to 3). <br> <strong>clicks.txt</strong> contains a list of the classified echolocalization emission events detected by an algorithm specifically developed for this aim, where: the first column reports the starting time offset the second column reports the end-time offset the third column contains the classification in classes of 5ms (f.e. class 5 means 20-25 ms class range, class 10 means 45-50 ms range) <br> <br> These txt files can be directly used with Audacity software to mark the vocalization contained in the raw wave file stored in DATA 1 as described in this tutorial. <br> <strong>results.xlsx</strong> reports the results of an algorithm-based analysis of 99 acoustic segments, achieved splitting the raw audio signal in 98 one-minute-length parts and one 20-second-lenght final part. This analysis was performed in order to detect and classify high-frequency dolphin’s emissions. The csv file includes the following information for each acoustic segment: the segment number; the total number of considered peaks; the number of detected click trains; the minimum ICI value of the detected trains; the maximum ICI value of the detected trains; the average ICI value of the detected trains; the number of observed whistles, if present in the segment; the presence in the relative segment of multiple whistles (Yes/No). <br>