Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Yuliang, Togneri, Roberto, Defeng, Huang
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2408.14771
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909316026466304
author Zhang, Yuliang
Togneri, Roberto
Defeng
Huang
author_facet Zhang, Yuliang
Togneri, Roberto
Defeng
Huang
contents This study explores the critical but underexamined impact of label noise on Sound Event Detection (SED), which requires both sound identification and precise temporal localization. We categorize label noise into deletion, insertion, substitution, and subjective types and systematically evaluate their effects on SED using synthetic and real-life datasets. Our analysis shows that deletion noise significantly degrades performance, while insertion noise is relatively benign. Moreover, loss functions effective against classification noise do not perform well for SED due to intra-class imbalance between foreground sound events and background sounds. We demonstrate that loss functions designed to address data imbalance in SED can effectively reduce the impact of noisy labels on system performance. For instance, halving the weight of background sounds in a synthetic dataset improved macro-F1 and micro-F1 scores by approximately $9\%$ with minimal Error Rate increase, with consistent results in real-life datasets. This research highlights the nuanced effects of noisy labels on SED systems and provides practical strategies to enhance model robustness, which are pivotal for both constructing new SED datasets and improving model performance, including efficient utilization of soft and crowdsourced labels.
format Preprint
id arxiv_https___arxiv_org_abs_2408_14771
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Impact of Noisy Labels on Sound Event Detection: Deletion Errors Are More Detrimental Than Insertion Errors
Zhang, Yuliang
Togneri, Roberto
Defeng
Huang
Audio and Speech Processing
This study explores the critical but underexamined impact of label noise on Sound Event Detection (SED), which requires both sound identification and precise temporal localization. We categorize label noise into deletion, insertion, substitution, and subjective types and systematically evaluate their effects on SED using synthetic and real-life datasets. Our analysis shows that deletion noise significantly degrades performance, while insertion noise is relatively benign. Moreover, loss functions effective against classification noise do not perform well for SED due to intra-class imbalance between foreground sound events and background sounds. We demonstrate that loss functions designed to address data imbalance in SED can effectively reduce the impact of noisy labels on system performance. For instance, halving the weight of background sounds in a synthetic dataset improved macro-F1 and micro-F1 scores by approximately $9\%$ with minimal Error Rate increase, with consistent results in real-life datasets. This research highlights the nuanced effects of noisy labels on SED systems and provides practical strategies to enhance model robustness, which are pivotal for both constructing new SED datasets and improving model performance, including efficient utilization of soft and crowdsourced labels.
title Impact of Noisy Labels on Sound Event Detection: Deletion Errors Are More Detrimental Than Insertion Errors
topic Audio and Speech Processing
url https://arxiv.org/abs/2408.14771