Flow-based Self-supervised Density Estimation for Anomalous Sound Detection
This work addresses incremental improvements in unsupervised anomaly detection for industrial machine sound monitoring systems.
The paper tackles the problem of anomalous sound detection in machine monitoring by improving Normalizing Flows to better distinguish target machine sounds from others of the same type, resulting in average AUC improvements of 4.6% with MAF and 5.8% with Glow over previous methods.
To develop a machine sound monitoring system, a method for detecting anomalous sound is proposed. Exact likelihood estimation using Normalizing Flows is a promising technique for unsupervised anomaly detection, but it can fail at out-of-distribution detection since the likelihood is affected by the smoothness of the data. To improve the detection performance, we train the model to assign higher likelihood to target machine sounds and lower likelihood to sounds from other machines of the same machine type. We demonstrate that this enables the model to incorporate a self-supervised classification-based approach. Experiments conducted using the DCASE 2020 Challenge Task2 dataset showed that the proposed method improves the AUC by 4.6% on average when using Masked Autoregressive Flow (MAF) and by 5.8% when using Glow, which is a significant improvement over the previous method.