Overcomplete Frame Thresholding for Acoustic Scene Analysis
This work addresses the challenge of applying optimal overcomplete representations in practical acoustic scene analysis, such as bird activity detection, but appears incremental as it focuses on thresholding within an existing framework.
The authors tackled the problem of leveraging overcomplete frames for real-world acoustic analysis tasks by deriving a generic thresholding scheme based on risk minimization, and validated it on a large-scale bird activity detection task using scattering networks with continuous wavelets.
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.