Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ASDKit: A Toolkit for Comprehensive Evaluation of Anomalous Sound Detection Methods

About

In this paper, we introduce ASDKit, a toolkit for anomalous sound detection (ASD) task. Our aim is to facilitate ASD research by providing an open-source framework that collects and carefully evaluates various ASD methods. First, ASDKit provides training and evaluation scripts for a wide range of ASD methods, all handled within a unified framework. For instance, it includes the autoencoder-based official DCASE baseline, representative discriminative methods, and self-supervised learning-based methods. Second, it supports comprehensive evaluation on the DCASE 2020--2024 datasets, enabling careful assessment of ASD performance, which is highly sensitive to factors such as datasets and random seeds. In our experiments, we re-evaluate various ASD methods using ASDKit and identify consistently effective techniques across multiple datasets and trials. We also demonstrate that ASDKit reproduces the state-of-the-art-level performance on the considered datasets.

Takuya Fujimura, Kevin Wilkinghoff, Keisuke Imoto, Tomoki Toda• 2025

Related benchmarks

TaskDatasetResultRank
Anomalous Sound DetectionDCASE 2020 (dev)
Official Performance Metric76.9
46
Anomalous Sound DetectionDCASE 2023 (dev)
Performance Metric62.3
17
Anomalous Sound DetectionDCASE 2023 (eval)
Official Performance Score62.6
17
Anomalous Sound DetectionDCASE 2020
Dataset-wise Harmonic Mean77.2
16
Anomalous Sound DetectionDCASE 2024
Dataset-wise Harmonic Mean56.7
16
Anomalous Sound DetectionDCASE 2023
Dataset-wise Harmonic Mean62.4
16
Anomalous Sound DetectionDCASE 2024 (eval)
Official Performance Metric55.7
16
Anomalous Sound DetectionDCASE 2020 (eval)
Official Performance Metric77.5
15
Anomalous Sound DetectionDCASE 2024 (dev)
Performance Score57.7
14
Anomalous Sound DetectionDCASE 2022
Dataset-wise Harmonic Mean60.9
12
Showing 10 of 12 rows

Other info

Follow for update