An Enhanced Audio Feature Tailored for Anomalous Sound Detection Based on Pre-trained Models

TL;DR

This paper introduces a novel audio feature tailored for anomalous sound detection that leverages pre-trained models and a filter bank design to improve detection accuracy by focusing on all frequency ranges and removing redundant noise.

Contribution

It proposes a new filter bank-based audio feature and a parameter-free enhancement method utilizing pre-trained models, advancing ASD performance and transfer learning capabilities.

Findings

Significant performance improvements on DCASE 2024 dataset

Effective removal of redundant noise in machine sounds

Enhanced detection of anomalies across all frequency ranges

Abstract

Anomalous Sound Detection (ASD) aims at identifying anomalous sounds from machines and has gained extensive research interests from both academia and industry. However, the uncertainty of anomaly location and much redundant information such as noise in machine sounds hinder the improvement of ASD system performance. This paper proposes a novel audio feature of filter banks with evenly distributed intervals, ensuring equal attention to all frequency ranges in the audio, which enhances the detection of anomalies in machine sounds. Moreover, based on pre-trained models, this paper presents a parameter-free feature enhancement approach to remove redundant information in machine audio. It is believed that this parameter-free strategy facilitates the effective transfer of universal knowledge from pre-trained tasks to the ASD task during model fine-tuning. Evaluation results on the Detection and Classification of Acoustic Scenes and Events (DCASE) 2024 Challenge dataset demonstrate significant improvements in ASD performance with our proposed methods.