Data for Generative AI-based data augmentation for improved bioacoustic classification in noisy environments

Data used for the paper Generative AI-based data augmentation for improved bioacoustic classification in noisy environments diffusion_samples.zip contains generated samples by the stable diffusion model used to supplement the training dataBirdnet_all_files_27class_images.zip contains all the spectrograms of 3 second clips annotated by BirdNET that come from the 27 bird species of interestmodels.zip contains model weights for the DDPM and VQAE modelsneal_images.zip contains spectrograms of the test data annotated by NEALpaper_tabular_data.zip contains the tabular data used, including those with BirdNET labels and embeddings and test data in .csv format, and train/test splits in .npz format