A Unified Loss Function to Tackle Inter-Class and Intra-Class Data Imbalance in Sound Event Detection
Yuliang Zhang, Roberto B. Togneri, Defeng David Huang · 2024
Data imbalance is an important issue in data-driven deep-learning methodologies. In sound event detection (SED), there are two types of data imbalance issues caused by the diverse time duration of sound events: the data imbalance between sound event classes (inter-class imbalance) and the active/inactive imbalance within the class (intra-class imbalance). In this paper, we propose a unified loss function (ULF), which adeptly addresses both the inter-class imbalance and intra-class imbalance simultaneously. Evaluation experiments substantiate that the ULF consistently yields superior and more stable performance compared to existing loss functions that singularly address either type of imbalance. Furthermore, the ULF loss also enhances the model's capacity to detect hard-to-detect sound events.