Raw Data for Few-Shot Bird Call Classification: An Automated Pipeline for Few-Shot Bird Call Classification, a Case Study with the Tooth-Billed Pigeon

This dataset consists of two separate collections used in our study titled "An Automated Pipeline for Few-Shot Bird Call Classification: A Case Study with the Tooth-Billed Pigeon."The first dataset includes two folders: train_data and test_data. It comprises bird calls from six species found in the Samoa region: Crimson-crowned Fruit Dove, Honeyeater, Pacific Imperial Pigeon, Samoan Starling, and Tooth-Billed Pigeon. Please note that we have not included audio for the Tooth-Billed Pigeon due to its endangered status. The test_data The folder contains a mix of calls from these species, used to validate our classification model.The second dataset includes us_train_data and us_test_data folders. It contains bird calls from six U.S. species: Carolina Wren, Eastern Towhee, Kentucky Warbler, Northern Cardinal, and White-Throated Sparrow. The us_test_data folder includes only calls from the Kentucky Warbler, which we used to evaluate the model's accuracy.