• D. Ansline Lidiya
• M. MohanaPriya
• M. BanuPriya

In contemporary times, the bird population is undergoing significant changes due to various factors such as human intervention, climate change, global warming, forest fires, and deforestation. This dynamic scenario necessitates innovative approaches for monitoring bird populations and understanding their behavior. Manual identification of different bird species is a labor-intensive task, prompting the development of an automatic bird identification system using machine learning algorithms. This work focuses on leveraging Convolutional Neural Networks (CNNs) for automatic bird species detection, comparing their performance with traditionally used classifiers such as SVM, Random Forest, and SMACPY. The primary objective is to streamline the identification process, enabling efficient monitoring of bird populations without the need for extensive time and effort. The proposed methodology involves the utilization of a dataset comprising vocalizations of diverse bird species. The input dataset undergoes preprocessing steps, including framing, silence removal, and reconstruction. Subsequently, a spectrogram is constructed, serving as input to a Convolutional Neural Network. The CNN undergoes modification, testing, and classification, ultimately generating an output that is compared with pre-trained data. Bird species are classified based on their distinctive features. This automated bird species identification system addresses the contemporary challenges associated with manual identification efforts. The integration of advanced technologies, particularly Convolutional Neural Networks, enhances the efficiency and accuracy of bird species classification. By automating the identification process, this work contributes to the ongoing efforts in biodiversity monitoring and conservation, offering a robust and time-effective solution for observing and understanding the dynamics of bird populations.

Machine learning, Automatic identification, Convolutional neural network, SVM, Random Forest, SMACPY Pre-process, Spectrogram, Classification.

"Automated Bird Species Identification using Audio Signal Processing and Neural Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 2, page no.g246-g249, February-2024, Available :http://www.jetir.org/papers/JETIR2402632.pdf

"Automated Bird Species Identification using Audio Signal Processing and Neural Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 2, page no. ppg246-g249, February-2024, Available at : http://www.jetir.org/papers/JETIR2402632.pdf