A Dual-Path Framework with Frequency-and-Time Excited Network for Anomalous Sound Detection

TL;DR

This paper introduces a dual-path framework with a Frequency-and-Time Excited Network for improved anomalous sound detection, leveraging frequency and temporal features to achieve state-of-the-art results on a benchmark dataset.

Contribution

The paper proposes a novel dual-path framework with FTE-Net and FTC-Encoder for enhanced anomaly detection by learning frequency and temporal patterns.

Findings

Achieves state-of-the-art performance on DCASE 2023 dataset.

Effectively captures frequency and temporal features for anomaly detection.

Provides visualizations demonstrating the model's feature learning capabilities.

Abstract

In contrast to human speech, machine-generated sounds of the same type often exhibit consistent frequency characteristics and discernible temporal periodicity. However, leveraging these dual attributes in anomaly detection remains relatively under-explored. In this paper, we propose an automated dual-path framework that learns prominent frequency and temporal patterns for diverse machine types. One pathway uses a novel Frequency-and-Time Excited Network (FTE-Net) to learn the salient features across frequency and time axes of the spectrogram. It incorporates a Frequency-and-Time Chunkwise Encoder (FTC-Encoder) and an excitation network. The other pathway uses a 1D convolutional network for utterance-level spectrum. Experimental results on the DCASE 2023 task 2 dataset show the state-of-the-art performance of our proposed method. Moreover, visualizations of the intermediate feature maps in the excitation network are provided to illustrate the effectiveness of our method.