Sound Event Detection in Urban Audio With Single and Multi-Rate PCEN

TL;DR

This paper explores the use of PCEN spectrograms for urban sound event detection, introducing Multi-Rate PCEN to improve cross-class performance in overlapping urban audio scenarios.

Contribution

It presents a novel Multi-Rate PCEN method that enhances cross-class sound event detection performance over traditional PCEN configurations.

Findings

MRPCEN improves cross-class detection accuracy

Per-class PCEN parameter tuning yields better results

MRPCEN outperforms single-rate PCEN in urban audio detection

Abstract

Recent literature has demonstrated that the use of per-channel energy normalization (PCEN), has significant performance improvements over traditional log-scaled mel-frequency spectrograms in acoustic sound event detection (SED) in a multi-class setting with overlapping events. However, the configuration of PCEN's parameters is sensitive to the recording environment, the characteristics of the class of events of interest, and the presence of multiple overlapping events. This leads to improvements on a class-by-class basis, but poor cross-class performance. In this article, we experiment using PCEN spectrograms as an alternative method for SED in urban audio using the UrbanSED dataset, demonstrating per-class improvements based on parameter configuration. Furthermore, we address cross-class performance with PCEN using a novel method, Multi-Rate PCEN (MRPCEN). We demonstrate cross-class SED performance with MRPCEN, demonstrating improvements to cross-class performance compared to traditional single-rate PCEN.