Enhanced Forest Sound Classification Dataset: EFSC-24

Forest sound classification is essential for monitoring forest ecosystems and promoting sustainable forest management. This paper introduces an Enhanced Forest Sound Classification Dataset 2024 (EFSC-24) and proposes a novel stacking model to classify the forest sounds. The proposed model's performance is evaluated and compared with benchmark algorithms such as K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Random Forest, and XGBoost. for both FSC-22 and EFSC-24. The comparison of proposed stacking model with benchmark models exhibits performance gains of up to 58%.