Abstract: This article presents an enhanced approach for bird species classification from their recorded audio signals. Observing that textures of syllables in audio spectrograms have noticeable ...

The classification of bird species is of significant importance in the field of ornithology, as it plays an important role in assessing and monitoring environmental dynamics, including habitat ...

This project applies deep learning to classify bird species from audio recordings. By converting audio to spectrograms and training convolutional neural networks ...

Please cite our paper (s) if you find this repository useful. The first paper proposes the Audio Spectrogram Transformer while the second paper describes the training pipeline that we applied on AST ...

Over the course of any given year, more than 200 bird species live just west of Manhattan's spires in the Hackensack Meadowlands. Scientists have known that for a while, with the Meadowlands serving ...