Data Quantification and Database Establishment

Import the raw data of sound into Adobe Audition 3.0 or Avisoft-SASLab Pro sound analysis software, resampled (Sample size: 44100 Hz; Window size: 1024 points), and then saved separately in WAV format. High-quality waveforms and sonograms were selected to measure characteristics of Hainan gibbons’ calls, to analyze the differences in acoustic indexes between individuals, and to build a database of Hainan gibbon sound patterns on an individual basis. Then, perform individual sound recognition using the implemented sound recognition model. Finally, the effectiveness of the sound acquisition is evaluated, and the accuracy of the sound recognition is assessed. Among them, the evaluation of the sound recognition effect is done mainly by comparing with the field research and other sound monitoring results.

Based on the acquired time-frequency domain characteristics of Hainan gibbons, the parameters used for automatic recognition were determined in conjunction with the vocal database. The selected time-frequency parameters were imported into the automatic recognition software and the developed algorithm program to automatically identify and extract Hainan gibbon calls from the recordings. Information such as the number of gibbons that may be present in the sound data is evaluated by different clustering and discriminative methods.

The fully-automated acoustic monitoring equipment is of vital use for data processing in this project. The transmitted sound data is automatically stored in Huawei cloud space. Once the Hainan biodiversity sound pattern Huawei cloud database be established, individual sound recognition could be realized.