First-Shot Unsupervised Anomalous Sound Detection With Unknown Anomalies Estimated by Metadata-Assisted Audio Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Hejing, Zhu, Qiaoxi, Guan, Jian, Liu, Haohe, Xiao, Feiyang, Tian, Jiantong, Mei, Xinhao, Liu, Xubo, Wang, Wenwu
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910360497291264
author Zhang, Hejing
Zhu, Qiaoxi
Guan, Jian
Liu, Haohe
Xiao, Feiyang
Tian, Jiantong
Mei, Xinhao
Liu, Xubo
Wang, Wenwu
author_facet Zhang, Hejing
Zhu, Qiaoxi
Guan, Jian
Liu, Haohe
Xiao, Feiyang
Tian, Jiantong
Mei, Xinhao
Liu, Xubo
Wang, Wenwu
contents First-shot (FS) unsupervised anomalous sound detection (ASD) is a brand-new task introduced in DCASE 2023 Challenge Task 2, where the anomalous sounds for the target machine types are unseen in training. Existing methods often rely on the availability of normal and abnormal sound data from the target machines. However, due to the lack of anomalous sound data for the target machine types, it becomes challenging when adapting the existing ASD methods to the first-shot task. In this paper, we propose a new framework for the first-shot unsupervised ASD, where metadata-assisted audio generation is used to estimate unknown anomalies, by utilising the available machine information (i.e., metadata and sound data) to fine-tune a text-to-audio generation model for generating the anomalous sounds that contain unique acoustic characteristics accounting for each different machine type. We then use the method of Time-Weighted Frequency domain audio Representation with Gaussian Mixture Model (TWFR-GMM) as the backbone to achieve the first-shot unsupervised ASD. Our proposed FS-TWFR-GMM method achieves competitive performance amongst top systems in DCASE 2023 Challenge Task 2, while requiring only 1% model parameters for detection, as validated in our experiments.
format Preprint
id arxiv_https___arxiv_org_abs_2310_14173
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle First-Shot Unsupervised Anomalous Sound Detection With Unknown Anomalies Estimated by Metadata-Assisted Audio Generation
Zhang, Hejing
Zhu, Qiaoxi
Guan, Jian
Liu, Haohe
Xiao, Feiyang
Tian, Jiantong
Mei, Xinhao
Liu, Xubo
Wang, Wenwu
Sound
Audio and Speech Processing
First-shot (FS) unsupervised anomalous sound detection (ASD) is a brand-new task introduced in DCASE 2023 Challenge Task 2, where the anomalous sounds for the target machine types are unseen in training. Existing methods often rely on the availability of normal and abnormal sound data from the target machines. However, due to the lack of anomalous sound data for the target machine types, it becomes challenging when adapting the existing ASD methods to the first-shot task. In this paper, we propose a new framework for the first-shot unsupervised ASD, where metadata-assisted audio generation is used to estimate unknown anomalies, by utilising the available machine information (i.e., metadata and sound data) to fine-tune a text-to-audio generation model for generating the anomalous sounds that contain unique acoustic characteristics accounting for each different machine type. We then use the method of Time-Weighted Frequency domain audio Representation with Gaussian Mixture Model (TWFR-GMM) as the backbone to achieve the first-shot unsupervised ASD. Our proposed FS-TWFR-GMM method achieves competitive performance amongst top systems in DCASE 2023 Challenge Task 2, while requiring only 1% model parameters for detection, as validated in our experiments.
title First-Shot Unsupervised Anomalous Sound Detection With Unknown Anomalies Estimated by Metadata-Assisted Audio Generation
topic Sound
Audio and Speech Processing
url https://arxiv.org/abs/2310.14173