ACOUSTIC IDENTIFICATION OF BIRD SPECIES BASED ON NATURAL LANGUAGE PROCESSING METHODOLOGY IN NON-STATIONARY ENVIRONMENT

Abstract: Methodologies for their proof of identity have been researched, and an automated system for bird species reputation has been created. Invariably trying to identify bird calls without human intervention has proven to be a challenging and time-consuming task for extensive research in ornithology's taxonomy and various other subfields. An identity process at the level is hired for this venture. The first step involved constructing a perfect dataset with all of the sound recordings from different chicken species. The final step involved applying a variety of sound preprocessing techniques to the audio clips, including-emphasis, framing, removal of peace, and re-construction. For each and every constructed audio clip, spectrograms were produced. The second process entails establishing a neural community with the spectrograms as its input. based

Keywords: Deep Learning, Neural Networks, Image Processing, Convolution Neural Network