Self Training and Ensembling Frequency Dependent Networks with Coarse Prediction Pooling and Sound Event Bounding Boxes

TL;DR

To tackle sound event detection (SED), this work proposes frequency dependent networks (FreDNets), which heavily leverage frequency-dependent methods, and applies frequency warping and FilterAugment, which are frequency-dependent data augmentation methods.

Abstract

To tackle sound event detection (SED), we propose frequency dependent networks (FreDNets), which heavily leverage frequency-dependent methods. We apply frequency warping and FilterAugment, which are frequency-dependent data augmentation methods. The model architecture consists of 3 branches: audio teacher-student transformer (ATST) branch, BEATs branch and CNN branch including either partial dilated frequency dynamic convolution (PDFD conv) or squeeze-and-Excitation (SE) with time-frame frequency-wise SE (tfwSE). To train MAESTRO labels with coarse temporal resolution, we applied max pooling on prediction for the MAESTRO dataset. Using best ensemble model, we applied self training to obtain pseudo label from DESED weak set, unlabeled set and AudioSet. AudioSet pseudo labels, filtered to focus on high-confidence labels, are used to train on DESED dataset only. We used change-detection-based sound event bounding boxes (cSEBBs) as post processing for ensemble models on self training and submission models. The resulting FreDNet was ranked 2nd in DCASE 2024 Challenge Task 4.

Figures (2)

• Figure 1: An illustration of framework for training and self training FreDNets.
• Figure 2: An illustration of partial dilated frequency dynamic convolution (PDFD conv). It involves a dynamic DFD conv branches and a static 2D convolution branch.