Loading…
Acoustic scene classification based on three-dimensional multi-channel feature-correlated deep learning networks
As an effective approach to perceive environments, acoustic scene classification (ASC) has received considerable attention in the past few years. Generally, ASC is deemed a challenging task due to subtle differences between various classes of environmental sounds. In this paper, we propose a novel a...
Saved in:
Published in: | Scientific reports 2022-08, Vol.12 (1), p.13730-13730, Article 13730 |
---|---|
Main Authors: | , , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | cdi_FETCH-LOGICAL-c583t-a31b3e3e68ba27e0615f956b8aac6d6d833b29f8a70b91e56a80072e0611d39b3 |
---|---|
cites | cdi_FETCH-LOGICAL-c583t-a31b3e3e68ba27e0615f956b8aac6d6d833b29f8a70b91e56a80072e0611d39b3 |
container_end_page | 13730 |
container_issue | 1 |
container_start_page | 13730 |
container_title | Scientific reports |
container_volume | 12 |
creator | Qu, Yuanyuan Li, Xuesheng Qin, Zhiliang Lu, Qidong |
description | As an effective approach to perceive environments, acoustic scene classification (ASC) has received considerable attention in the past few years. Generally, ASC is deemed a challenging task due to subtle differences between various classes of environmental sounds. In this paper, we propose a novel approach to perform accurate classification based on the aggregation of spatial–temporal features extracted from a multi-branch three-dimensional (3D) convolution neural network (CNN) model. The novelties of this paper are as follows. First, we form multiple frequency-domain representations of signals by fully utilizing expert knowledge on acoustics and discrete wavelet transformations (DWT). Secondly, we propose a novel 3D CNN architecture featuring residual connections and squeeze-and-excitation attentions (3D-SE-ResNet) to effectively capture both long-term and short-term correlations inherent in environmental sounds. Thirdly, an auxiliary supervised branch based on the chromatogram of the original signal is incorporated in the proposed architecture to alleviate overfitting risks by providing supplementary information to the model. The performance of the proposed multi-input multi-feature 3D-CNN architecture is numerically evaluated on a typical large-scale dataset in the 2019 IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE 2019) and is shown to obtain noticeable performance gains over the state-of-the-art methods in the literature. |
doi_str_mv | 10.1038/s41598-022-17863-z |
format | article |
fullrecord | <record><control><sourceid>proquest_doaj_</sourceid><recordid>TN_cdi_doaj_primary_oai_doaj_org_article_4c3bc95273b0483290e3787a76b49680</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><doaj_id>oai_doaj_org_article_4c3bc95273b0483290e3787a76b49680</doaj_id><sourcerecordid>2701360096</sourcerecordid><originalsourceid>FETCH-LOGICAL-c583t-a31b3e3e68ba27e0615f956b8aac6d6d833b29f8a70b91e56a80072e0611d39b3</originalsourceid><addsrcrecordid>eNp9kktv1DAUhSMEolXpH2AViQ2bgB-JHxukquJRqRIbWFvXzs2MB4892Amo_fV4JhVQFnhjy_ecT77Xp2leUvKGEq7elp4OWnWEsY5KJXh3_6Q5Z6QfOsYZe_rX-ay5LGVH6hqY7ql-3pzxQQtGGD1vDlcuLWX2ri0OI7YuQCl-8g5mn2JroeDY1sO8zYjd6PcYSy1AaPdLmH3nthAjhnZCmJeMnUs5Y4C5ukbEQxsQcvRx00acf6b8rbxonk0QCl4-7BfN1w_vv1x_6m4_f7y5vrrt3KD43AGnliNHoSwwiUTQYdKDsArAiVGMinPL9KRAEqspDgIUIZIdhXTk2vKL5mbljgl25pD9HvKdSeDN6SLljYFc-w5oeset0wOT3JJecaYJcqkkSGF7LRSprHcr67DYPY51UHOG8Aj6uBL91mzSD6O57IUUFfD6AZDT9wXLbPa-zjsEiFjHb5isn3FsYKjSV_9Id2nJdeAnFeWCEH0EslXlciol4_T7MZSYYz7Mmg9T82FO-TD31cRXU6niuMH8B_0f1y8mIr3a</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2701360096</pqid></control><display><type>article</type><title>Acoustic scene classification based on three-dimensional multi-channel feature-correlated deep learning networks</title><source>PubMed Central Free</source><source>Free Full-Text Journals in Chemistry</source><source>Springer Nature - nature.com Journals - Fully Open Access</source><source>ProQuest Publicly Available Content database</source><creator>Qu, Yuanyuan ; Li, Xuesheng ; Qin, Zhiliang ; Lu, Qidong</creator><creatorcontrib>Qu, Yuanyuan ; Li, Xuesheng ; Qin, Zhiliang ; Lu, Qidong</creatorcontrib><description>As an effective approach to perceive environments, acoustic scene classification (ASC) has received considerable attention in the past few years. Generally, ASC is deemed a challenging task due to subtle differences between various classes of environmental sounds. In this paper, we propose a novel approach to perform accurate classification based on the aggregation of spatial–temporal features extracted from a multi-branch three-dimensional (3D) convolution neural network (CNN) model. The novelties of this paper are as follows. First, we form multiple frequency-domain representations of signals by fully utilizing expert knowledge on acoustics and discrete wavelet transformations (DWT). Secondly, we propose a novel 3D CNN architecture featuring residual connections and squeeze-and-excitation attentions (3D-SE-ResNet) to effectively capture both long-term and short-term correlations inherent in environmental sounds. Thirdly, an auxiliary supervised branch based on the chromatogram of the original signal is incorporated in the proposed architecture to alleviate overfitting risks by providing supplementary information to the model. The performance of the proposed multi-input multi-feature 3D-CNN architecture is numerically evaluated on a typical large-scale dataset in the 2019 IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE 2019) and is shown to obtain noticeable performance gains over the state-of-the-art methods in the literature.</description><identifier>ISSN: 2045-2322</identifier><identifier>EISSN: 2045-2322</identifier><identifier>DOI: 10.1038/s41598-022-17863-z</identifier><identifier>PMID: 35962021</identifier><language>eng</language><publisher>London: Nature Publishing Group UK</publisher><subject>639/166/987 ; 704/172/4081 ; Acoustics ; Classification ; Deep learning ; Humanities and Social Sciences ; multidisciplinary ; Neural networks ; Science ; Science (multidisciplinary) ; Temporal variations</subject><ispartof>Scientific reports, 2022-08, Vol.12 (1), p.13730-13730, Article 13730</ispartof><rights>The Author(s) 2022</rights><rights>The Author(s) 2022. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c583t-a31b3e3e68ba27e0615f956b8aac6d6d833b29f8a70b91e56a80072e0611d39b3</citedby><cites>FETCH-LOGICAL-c583t-a31b3e3e68ba27e0615f956b8aac6d6d833b29f8a70b91e56a80072e0611d39b3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.proquest.com/docview/2701360096/fulltextPDF?pq-origsite=primo$$EPDF$$P50$$Gproquest$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://www.proquest.com/docview/2701360096?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>230,314,727,780,784,885,25753,27924,27925,37012,37013,44590,53791,53793,75126</link.rule.ids></links><search><creatorcontrib>Qu, Yuanyuan</creatorcontrib><creatorcontrib>Li, Xuesheng</creatorcontrib><creatorcontrib>Qin, Zhiliang</creatorcontrib><creatorcontrib>Lu, Qidong</creatorcontrib><title>Acoustic scene classification based on three-dimensional multi-channel feature-correlated deep learning networks</title><title>Scientific reports</title><addtitle>Sci Rep</addtitle><description>As an effective approach to perceive environments, acoustic scene classification (ASC) has received considerable attention in the past few years. Generally, ASC is deemed a challenging task due to subtle differences between various classes of environmental sounds. In this paper, we propose a novel approach to perform accurate classification based on the aggregation of spatial–temporal features extracted from a multi-branch three-dimensional (3D) convolution neural network (CNN) model. The novelties of this paper are as follows. First, we form multiple frequency-domain representations of signals by fully utilizing expert knowledge on acoustics and discrete wavelet transformations (DWT). Secondly, we propose a novel 3D CNN architecture featuring residual connections and squeeze-and-excitation attentions (3D-SE-ResNet) to effectively capture both long-term and short-term correlations inherent in environmental sounds. Thirdly, an auxiliary supervised branch based on the chromatogram of the original signal is incorporated in the proposed architecture to alleviate overfitting risks by providing supplementary information to the model. The performance of the proposed multi-input multi-feature 3D-CNN architecture is numerically evaluated on a typical large-scale dataset in the 2019 IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE 2019) and is shown to obtain noticeable performance gains over the state-of-the-art methods in the literature.</description><subject>639/166/987</subject><subject>704/172/4081</subject><subject>Acoustics</subject><subject>Classification</subject><subject>Deep learning</subject><subject>Humanities and Social Sciences</subject><subject>multidisciplinary</subject><subject>Neural networks</subject><subject>Science</subject><subject>Science (multidisciplinary)</subject><subject>Temporal variations</subject><issn>2045-2322</issn><issn>2045-2322</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><sourceid>DOA</sourceid><recordid>eNp9kktv1DAUhSMEolXpH2AViQ2bgB-JHxukquJRqRIbWFvXzs2MB4892Amo_fV4JhVQFnhjy_ecT77Xp2leUvKGEq7elp4OWnWEsY5KJXh3_6Q5Z6QfOsYZe_rX-ay5LGVH6hqY7ql-3pzxQQtGGD1vDlcuLWX2ri0OI7YuQCl-8g5mn2JroeDY1sO8zYjd6PcYSy1AaPdLmH3nthAjhnZCmJeMnUs5Y4C5ukbEQxsQcvRx00acf6b8rbxonk0QCl4-7BfN1w_vv1x_6m4_f7y5vrrt3KD43AGnliNHoSwwiUTQYdKDsArAiVGMinPL9KRAEqspDgIUIZIdhXTk2vKL5mbljgl25pD9HvKdSeDN6SLljYFc-w5oeset0wOT3JJecaYJcqkkSGF7LRSprHcr67DYPY51UHOG8Aj6uBL91mzSD6O57IUUFfD6AZDT9wXLbPa-zjsEiFjHb5isn3FsYKjSV_9Id2nJdeAnFeWCEH0EslXlciol4_T7MZSYYz7Mmg9T82FO-TD31cRXU6niuMH8B_0f1y8mIr3a</recordid><startdate>20220812</startdate><enddate>20220812</enddate><creator>Qu, Yuanyuan</creator><creator>Li, Xuesheng</creator><creator>Qin, Zhiliang</creator><creator>Lu, Qidong</creator><general>Nature Publishing Group UK</general><general>Nature Publishing Group</general><general>Nature Portfolio</general><scope>C6C</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>3V.</scope><scope>7X7</scope><scope>7XB</scope><scope>88A</scope><scope>88E</scope><scope>88I</scope><scope>8FE</scope><scope>8FH</scope><scope>8FI</scope><scope>8FJ</scope><scope>8FK</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BBNVY</scope><scope>BENPR</scope><scope>BHPHI</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>FYUFA</scope><scope>GHDGH</scope><scope>GNUQQ</scope><scope>HCIFZ</scope><scope>K9.</scope><scope>LK8</scope><scope>M0S</scope><scope>M1P</scope><scope>M2P</scope><scope>M7P</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>Q9U</scope><scope>7X8</scope><scope>5PM</scope><scope>DOA</scope></search><sort><creationdate>20220812</creationdate><title>Acoustic scene classification based on three-dimensional multi-channel feature-correlated deep learning networks</title><author>Qu, Yuanyuan ; Li, Xuesheng ; Qin, Zhiliang ; Lu, Qidong</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c583t-a31b3e3e68ba27e0615f956b8aac6d6d833b29f8a70b91e56a80072e0611d39b3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>639/166/987</topic><topic>704/172/4081</topic><topic>Acoustics</topic><topic>Classification</topic><topic>Deep learning</topic><topic>Humanities and Social Sciences</topic><topic>multidisciplinary</topic><topic>Neural networks</topic><topic>Science</topic><topic>Science (multidisciplinary)</topic><topic>Temporal variations</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Qu, Yuanyuan</creatorcontrib><creatorcontrib>Li, Xuesheng</creatorcontrib><creatorcontrib>Qin, Zhiliang</creatorcontrib><creatorcontrib>Lu, Qidong</creatorcontrib><collection>SpringerOpen</collection><collection>CrossRef</collection><collection>ProQuest Central (Corporate)</collection><collection>ProQuest_Health & Medical Collection</collection><collection>ProQuest Central (purchase pre-March 2016)</collection><collection>Biology Database (Alumni Edition)</collection><collection>Medical Database (Alumni Edition)</collection><collection>Science Database (Alumni Edition)</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Natural Science Collection</collection><collection>Hospital Premium Collection</collection><collection>Hospital Premium Collection (Alumni Edition)</collection><collection>ProQuest Central (Alumni) (purchase pre-March 2016)</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>Biological Science Collection</collection><collection>ProQuest Central</collection><collection>Natural Science Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>Health Research Premium Collection</collection><collection>Health Research Premium Collection (Alumni)</collection><collection>ProQuest Central Student</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Health & Medical Complete (Alumni)</collection><collection>ProQuest Biological Science Collection</collection><collection>Health & Medical Collection (Alumni Edition)</collection><collection>PML(ProQuest Medical Library)</collection><collection>Science Database</collection><collection>ProQuest Biological Science Journals</collection><collection>ProQuest Publicly Available Content database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central Basic</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><collection>DOAJ Directory of Open Access Journals</collection><jtitle>Scientific reports</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Qu, Yuanyuan</au><au>Li, Xuesheng</au><au>Qin, Zhiliang</au><au>Lu, Qidong</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Acoustic scene classification based on three-dimensional multi-channel feature-correlated deep learning networks</atitle><jtitle>Scientific reports</jtitle><stitle>Sci Rep</stitle><date>2022-08-12</date><risdate>2022</risdate><volume>12</volume><issue>1</issue><spage>13730</spage><epage>13730</epage><pages>13730-13730</pages><artnum>13730</artnum><issn>2045-2322</issn><eissn>2045-2322</eissn><abstract>As an effective approach to perceive environments, acoustic scene classification (ASC) has received considerable attention in the past few years. Generally, ASC is deemed a challenging task due to subtle differences between various classes of environmental sounds. In this paper, we propose a novel approach to perform accurate classification based on the aggregation of spatial–temporal features extracted from a multi-branch three-dimensional (3D) convolution neural network (CNN) model. The novelties of this paper are as follows. First, we form multiple frequency-domain representations of signals by fully utilizing expert knowledge on acoustics and discrete wavelet transformations (DWT). Secondly, we propose a novel 3D CNN architecture featuring residual connections and squeeze-and-excitation attentions (3D-SE-ResNet) to effectively capture both long-term and short-term correlations inherent in environmental sounds. Thirdly, an auxiliary supervised branch based on the chromatogram of the original signal is incorporated in the proposed architecture to alleviate overfitting risks by providing supplementary information to the model. The performance of the proposed multi-input multi-feature 3D-CNN architecture is numerically evaluated on a typical large-scale dataset in the 2019 IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE 2019) and is shown to obtain noticeable performance gains over the state-of-the-art methods in the literature.</abstract><cop>London</cop><pub>Nature Publishing Group UK</pub><pmid>35962021</pmid><doi>10.1038/s41598-022-17863-z</doi><tpages>1</tpages><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 2045-2322 |
ispartof | Scientific reports, 2022-08, Vol.12 (1), p.13730-13730, Article 13730 |
issn | 2045-2322 2045-2322 |
language | eng |
recordid | cdi_doaj_primary_oai_doaj_org_article_4c3bc95273b0483290e3787a76b49680 |
source | PubMed Central Free; Free Full-Text Journals in Chemistry; Springer Nature - nature.com Journals - Fully Open Access; ProQuest Publicly Available Content database |
subjects | 639/166/987 704/172/4081 Acoustics Classification Deep learning Humanities and Social Sciences multidisciplinary Neural networks Science Science (multidisciplinary) Temporal variations |
title | Acoustic scene classification based on three-dimensional multi-channel feature-correlated deep learning networks |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T12%3A12%3A49IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_doaj_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Acoustic%20scene%20classification%20based%20on%20three-dimensional%20multi-channel%20feature-correlated%20deep%20learning%20networks&rft.jtitle=Scientific%20reports&rft.au=Qu,%20Yuanyuan&rft.date=2022-08-12&rft.volume=12&rft.issue=1&rft.spage=13730&rft.epage=13730&rft.pages=13730-13730&rft.artnum=13730&rft.issn=2045-2322&rft.eissn=2045-2322&rft_id=info:doi/10.1038/s41598-022-17863-z&rft_dat=%3Cproquest_doaj_%3E2701360096%3C/proquest_doaj_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c583t-a31b3e3e68ba27e0615f956b8aac6d6d833b29f8a70b91e56a80072e0611d39b3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2701360096&rft_id=info:pmid/35962021&rfr_iscdi=true |