Forest quality assessment based on bird sound recognition using convolutional neural networks

Deforestation in Indonesia is a status that quite alarming. From year to year, deforestation still happening. The decline fauna and the diminishing biodiversity are greatly affected by deforestation. This paper proposes bioacoustics-based forest quality assessment tool using Nvidia Jetson Nano convolutional neural networks (CNN). device, named GamaDet, portable physical product based on microprocessor equipped with microphone record sounds of birds display results their analysis. In addition, Google Collaboratorybased GamaNet digital also proposed. requires recording audio files be further analyzed into index. Testing for 60 seconds at an arboretum showed both products could work well. GamaDet takes 370 seconds, while 70 process data index list detected birds.