• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利:CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验

不同声学环境下的特征提取与机器学习方法的比较研究。

A Comparative Survey of Feature Extraction and Machine Learning Methods in Diverse Acoustic Environments.

机构信息

Grup de Recerca en Tecnologies Mèdia (GTM), La Salle-URL, c/Quatre Camins, 30, 08022 Barcelona, Spain.

出版信息

Sensors (Basel). 2021 Feb 11;21(4):1274. doi: 10.3390/s21041274.

DOI:10.3390/s21041274
PMID:33670096
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC7916834/
Abstract

Acoustic event detection and analysis has been widely developed in the last few years for its valuable application in monitoring elderly or dependant people, for surveillance issues, for multimedia retrieval, or even for biodiversity metrics in natural environments. For this purpose, sound source identification is a key issue to give a smart technological answer to all the aforementioned applications. Diverse types of sounds and variate environments, together with a number of challenges in terms of application, widen the choice of artificial intelligence algorithm proposal. This paper presents a comparative study on combining several feature extraction algorithms (Mel Frequency Cepstrum Coefficients (MFCC), Gammatone Cepstrum Coefficients (GTCC), and Narrow Band (NB)) with a group of machine learning algorithms (-Nearest Neighbor (kNN), Neural Networks (NN), and Gaussian Mixture Model (GMM)), tested over five different acoustic environments. This work has the goal of detailing a best practice method and evaluate the reliability of this general-purpose algorithm for all the classes. Preliminary results show that most of the combinations of feature extraction and machine learning present acceptable results in most of the described corpora. Nevertheless, there is a combination that outperforms the others: the use of GTCC together with kNN, and its results are further analyzed for all the corpora.

摘要

在过去的几年中,声学事件检测和分析得到了广泛的发展,因为它在监测老年人或依赖者、监控问题、多媒体检索,甚至在自然环境中的生物多样性指标方面都有很有价值的应用。为此,声源识别是为所有上述应用提供智能技术答案的关键问题。不同类型的声音和多变的环境,以及在应用方面的一些挑战,扩大了人工智能算法建议的选择范围。本文对几种特征提取算法(梅尔频率倒谱系数 (MFCC)、Gamma 频率倒谱系数 (GTCC) 和窄带 (NB))与一组机器学习算法 (-最近邻 (kNN)、神经网络 (NN) 和高斯混合模型 (GMM))进行了比较研究,在五个不同的声学环境中进行了测试。这项工作的目的是详细说明一种最佳实践方法,并评估这种通用算法对所有类别的可靠性。初步结果表明,在大多数描述的语料库中,特征提取和机器学习的大多数组合都能得到可接受的结果。然而,有一种组合优于其他组合:使用 GTCC 结合 kNN,并且对所有语料库进一步分析了其结果。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/9b3a9bc2c40a/sensors-21-01274-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/2b4e89d0b0b8/sensors-21-01274-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/39a4060375ad/sensors-21-01274-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/82dcfab3faa1/sensors-21-01274-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/697405a07b5f/sensors-21-01274-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/8cbaddd770f3/sensors-21-01274-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/9b3a9bc2c40a/sensors-21-01274-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/2b4e89d0b0b8/sensors-21-01274-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/39a4060375ad/sensors-21-01274-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/82dcfab3faa1/sensors-21-01274-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/697405a07b5f/sensors-21-01274-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/8cbaddd770f3/sensors-21-01274-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/9336/7916834/9b3a9bc2c40a/sensors-21-01274-g006.jpg

相似文献

1
A Comparative Survey of Feature Extraction and Machine Learning Methods in Diverse Acoustic Environments.不同声学环境下的特征提取与机器学习方法的比较研究。
Sensors (Basel). 2021 Feb 11;21(4):1274. doi: 10.3390/s21041274.
2
On effective cognitive state classification using novel feature extraction strategies.关于使用新型特征提取策略进行有效的认知状态分类
Cogn Neurodyn. 2021 Dec;15(6):1125-1155. doi: 10.1007/s11571-021-09688-9. Epub 2021 Jun 22.
3
A comparative study of the SVM and K-nn machine learning algorithms for the diagnosis of respiratory pathologies using pulmonary acoustic signals.基于肺部声信号的呼吸病理诊断的支持向量机(SVM)和 K-近邻(K-nn)机器学习算法的比较研究。
BMC Bioinformatics. 2014 Jun 27;15:223. doi: 10.1186/1471-2105-15-223.
4
Narrow-band autocorrelation function features for the automatic recognition of acoustic environments.窄带自相关函数特征用于声环境的自动识别。
J Acoust Soc Am. 2013 Jul;134(1):880-90. doi: 10.1121/1.4807807.
5
Automated Parkinson's disease recognition based on statistical pooling method using acoustic features.基于统计池化方法的利用声学特征的帕金森病自动识别
Med Hypotheses. 2020 Feb;135:109483. doi: 10.1016/j.mehy.2019.109483. Epub 2019 Nov 11.
6
A Novel Machine Learning-Based Methodology for Tool Wear Prediction Using Acoustic Emission Signals.基于声发射信号的刀具磨损预测新型机器学习方法。
Sensors (Basel). 2021 Sep 6;21(17):5984. doi: 10.3390/s21175984.
7
Event Detection for Distributed Acoustic Sensing: Combining Knowledge-Based, Classical Machine Learning, and Deep Learning Approaches.分布式声学传感的事件检测:结合基于知识、经典机器学习和深度学习方法
Sensors (Basel). 2021 Nov 12;21(22):7527. doi: 10.3390/s21227527.
8
Identification of oestrus cows based on vocalisation characteristics and machine learning technique using a dual-channel-equipped acoustic tag.基于发声特征和使用双通道声学标签的机器学习技术识别发情奶牛。
Animal. 2023 Jun;17(6):100811. doi: 10.1016/j.animal.2023.100811. Epub 2023 Apr 14.
9
Comparative study of respiratory sounds classification methods based on cepstral analysis and artificial neural networks.基于倒谱分析和人工神经网络的呼吸音分类方法的比较研究。
Comput Biol Med. 2024 Mar;171:108190. doi: 10.1016/j.compbiomed.2024.108190. Epub 2024 Feb 20.
10
Diagnosis of COVID-19 via acoustic analysis and artificial intelligence by monitoring breath sounds on smartphones.通过在智能手机上监测呼吸声的声学分析和人工智能诊断 COVID-19。
J Biomed Inform. 2022 Jun;130:104078. doi: 10.1016/j.jbi.2022.104078. Epub 2022 Apr 27.

引用本文的文献

1
Real-Time Acoustic Scene Recognition for Elderly Daily Routines Using Edge-Based Deep Learning.基于边缘深度学习的老年人日常活动实时声学场景识别
Sensors (Basel). 2025 Mar 12;25(6):1746. doi: 10.3390/s25061746.
2
Enhancing smart home environments: a novel pattern recognition approach to ambient acoustic event detection and localization.增强智能家居环境:一种用于环境声学事件检测与定位的新型模式识别方法。
Front Big Data. 2025 Jan 23;7:1419562. doi: 10.3389/fdata.2024.1419562. eCollection 2024.
3
BSN-ESC: A Big-Small Network-Based Environmental Sound Classification Method for AIoT Applications.

本文引用的文献

1
A WASN-Based Suburban Dataset for Anomalous Noise Event Detection on Dynamic Road-Traffic Noise Mapping.一种用于动态道路交通噪声映射中异常噪声事件检测的基于WASN的郊区数据集。
Sensors (Basel). 2019 May 30;19(11):2480. doi: 10.3390/s19112480.
2
Real-Time Distributed Architecture for Remote Acoustic Elderly Monitoring in Residential-Scale Ambient Assisted Living Scenarios.用于住宅规模的安养环境中远程声学老年人监护的实时分布式架构。
Sensors (Basel). 2018 Aug 1;18(8):2492. doi: 10.3390/s18082492.
3
An Anomalous Noise Events Detector for Dynamic Road Traffic Noise Mapping in Real-Life Urban and Suburban Environments.
BSN-ESC:一种用于物联网应用的基于大小网络的环境声音分类方法。
Sensors (Basel). 2023 Jul 28;23(15):6767. doi: 10.3390/s23156767.
4
Analysis and Acoustic Event Classification of Environmental Data Collected in a Citizen Science Project.在公民科学项目中收集的环境数据的分析和声学事件分类。
Int J Environ Res Public Health. 2023 Feb 19;20(4):3683. doi: 10.3390/ijerph20043683.
5
A Review of Automated Bioacoustics and General Acoustics Classification Research.自动生物声学与一般声学分类研究综述
Sensors (Basel). 2022 Oct 31;22(21):8361. doi: 10.3390/s22218361.
6
Machine Learning Smart System for Parkinson Disease Classification Using the Voice as a Biomarker.基于语音生物标志物的帕金森病分类机器学习智能系统
Healthc Inform Res. 2022 Jul;28(3):210-221. doi: 10.4258/hir.2022.28.3.210. Epub 2022 Jul 31.
7
Early Fault Diagnosis Method for Batch Process Based on Local Time Window Standardization and Trend Analysis.基于局部时间窗口标准化和趋势分析的批量过程早期故障诊断方法。
Sensors (Basel). 2021 Dec 2;21(23):8075. doi: 10.3390/s21238075.
8
Cluster Analysis of Urban Acoustic Environments on Barcelona Sensor Network Data.基于巴塞罗那传感器网络数据的城市声环境聚类分析。
Int J Environ Res Public Health. 2021 Aug 4;18(16):8271. doi: 10.3390/ijerph18168271.
9
Benchmarking Audio Signal Representation Techniques for Classification with Convolutional Neural Networks.基于卷积神经网络的音频信号分类技术的基准测试。
Sensors (Basel). 2021 May 14;21(10):3434. doi: 10.3390/s21103434.
一种用于现实城市和郊区环境中动态道路交通噪声映射的异常噪声事件检测器。
Sensors (Basel). 2017 Oct 12;17(10):2323. doi: 10.3390/s17102323.
4
An FPGA-Based WASN for Remote Real-Time Monitoring of Endangered Species: A Case Study on the Birdsong Recognition of Botaurus stellaris.基于 FPGA 的 WASN 用于濒危物种的远程实时监测:以东方白鹳鸟鸣识别为例
Sensors (Basel). 2017 Jun 8;17(6):1331. doi: 10.3390/s17061331.
5
homeSound: Real-Time Audio Event Detection Based on High Performance Computing for Behaviour and Surveillance Remote Monitoring.家庭声音:基于高性能计算的行为和监控远程监测的实时音频事件检测。
Sensors (Basel). 2017 Apr 13;17(4):854. doi: 10.3390/s17040854.
6
A survey on ambient-assisted living tools for older adults.老年人环境辅助生活工具调查。
IEEE J Biomed Health Inform. 2013 May;17(3):579-90. doi: 10.1109/jbhi.2012.2234129.
7
Narrow-band autocorrelation function features for the automatic recognition of acoustic environments.窄带自相关函数特征用于声环境的自动识别。
J Acoust Soc Am. 2013 Jul;134(1):880-90. doi: 10.1121/1.4807807.