Overcomplete Frame Thresholding for Acoustic Scene Analysis

TL;DR

This paper introduces a risk minimization-based thresholding scheme for overcomplete frames, enhancing acoustic scene analysis by improving representations in tasks like bird activity detection using scattering networks.

Contribution

It presents a novel, generic thresholding method for overcomplete frames tailored for analysis tasks, validated on real-world audio detection applications.

Findings

Effective in bird activity detection

Leverages scattering network architecture

Improves analysis of audio environments

Abstract

In this work, we derive a generic overcomplete frame thresholding scheme based on risk minimization. Overcomplete frames being favored for analysis tasks such as classification, regression or anomaly detection, we provide a way to leverage those optimal representations in real-world applications through the use of thresholding. We validate the method on a large scale bird activity detection task via the scattering network architecture performed by means of continuous wavelets, known for being an adequate dictionary in audio environments.