arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2310.03455eess.AScs.SD
收藏

性能与能量平衡:最先进声音事件检测系统的综合研究

Performance and energy balance: a comprehensive study of state-of-the-art sound event detection systems

Politecnico di Milano(米兰理工大学);Université de Lorraine(洛林大学)

Francesca Ronchini, Romain Serizel

AI总结:

该研究基于 DCASE 挑战赛提交结果,跨两年比较并详细分析最新 SED 系统,以揭示其性能演进与能耗、环境影响之间的关系。

摘要:

近年来,深度学习系统呈现出复杂性增加和能耗升高的令人担忧趋势。作为该领域的研究人员以及 Detection and Classification of Acoustic Scenes and Events 挑战赛其中一项任务的组织者,我们认识到解决数据驱动 SED 系统环境影响的重要性。本文基于挑战赛提交结果,提出一项聚焦 SED 系统的分析,包括过去两年的系统比较以及对今年 SED 系统的详细分析。通过本研究,我们旨在探索 SED 系统每年如何围绕其能效影响演进。

英文摘要:

In recent years, deep learning systems have shown a concerning trend toward increased complexity and higher energy consumption. As researchers in this domain and organizers of one of the Detection and Classification of Acoustic Scenes and Events challenges tasks, we recognize the importance of addressing the environmental impact of data-driven SED systems. In this paper, we propose an analysis focused on SED systems based on the challenge submissions. This includes a comparison across the past two years and a detailed analysis of this year's SED systems. Through this research, we aim to explore how the SED systems are evolving every year in relation to their energy efficiency implications.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑