{"id":"https://openalex.org/W7133543996","doi":"https://doi.org/10.1016/j.csl.2026.101967","title":"Exploring efficient attention strategies in conformer-based sound event detection","display_name":"Exploring efficient attention strategies in conformer-based sound event detection","publication_year":2026,"publication_date":"2026-03-04","ids":{"openalex":"https://openalex.org/W7133543996","doi":"https://doi.org/10.1016/j.csl.2026.101967"},"language":"en","primary_location":{"id":"doi:10.1016/j.csl.2026.101967","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.csl.2026.101967","pdf_url":null,"source":{"id":"https://openalex.org/S91252481","display_name":"Computer Speech & Language","issn_l":"0885-2308","issn":["0885-2308","1095-8363"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer Speech &amp; Language","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1016/j.csl.2026.101967","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5106528898","display_name":"Sara Barahona","orcid":"https://orcid.org/0009-0001-8519-0549"},"institutions":[{"id":"https://openalex.org/I63634437","display_name":"Universidad Aut\u00f3noma de Madrid","ror":"https://ror.org/01cby8j38","country_code":"ES","type":"education","lineage":["https://openalex.org/I63634437"]},{"id":"https://openalex.org/I88060688","display_name":"Universidad Polit\u00e9cnica de Madrid","ror":"https://ror.org/03n6nwv02","country_code":"ES","type":"education","lineage":["https://openalex.org/I88060688"]}],"countries":["ES"],"is_corresponding":true,"raw_author_name":"Sara Barahona","raw_affiliation_strings":["AUDIAS, Electronic and Communication Technology Department, Escuela Polit\u00e9cnica Superior, Universidad Aut\u00f3noma de Madrid, Av. Francisco Tom\u00e1s y Valiente, 11, Madrid, 28049, Spain"],"raw_orcid":"https://orcid.org/0009-0001-8519-0549","affiliations":[{"raw_affiliation_string":"AUDIAS, Electronic and Communication Technology Department, Escuela Polit\u00e9cnica Superior, Universidad Aut\u00f3noma de Madrid, Av. Francisco Tom\u00e1s y Valiente, 11, Madrid, 28049, Spain","institution_ids":["https://openalex.org/I63634437","https://openalex.org/I88060688"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078018833","display_name":"Juan Ignacio \u00c1lvarez-Trejos","orcid":"https://orcid.org/0000-0003-4757-3132"},"institutions":[{"id":"https://openalex.org/I63634437","display_name":"Universidad Aut\u00f3noma de Madrid","ror":"https://ror.org/01cby8j38","country_code":"ES","type":"education","lineage":["https://openalex.org/I63634437"]},{"id":"https://openalex.org/I88060688","display_name":"Universidad Polit\u00e9cnica de Madrid","ror":"https://ror.org/03n6nwv02","country_code":"ES","type":"education","lineage":["https://openalex.org/I88060688"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Juan Ignacio Alvarez-Trejos","raw_affiliation_strings":["AUDIAS, Electronic and Communication Technology Department, Escuela Polit\u00e9cnica Superior, Universidad Aut\u00f3noma de Madrid, Av. Francisco Tom\u00e1s y Valiente, 11, Madrid, 28049, Spain"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AUDIAS, Electronic and Communication Technology Department, Escuela Polit\u00e9cnica Superior, Universidad Aut\u00f3noma de Madrid, Av. Francisco Tom\u00e1s y Valiente, 11, Madrid, 28049, Spain","institution_ids":["https://openalex.org/I63634437","https://openalex.org/I88060688"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063156730","display_name":"Alicia Lozano-D\u00edez","orcid":"https://orcid.org/0000-0002-5918-8568"},"institutions":[{"id":"https://openalex.org/I63634437","display_name":"Universidad Aut\u00f3noma de Madrid","ror":"https://ror.org/01cby8j38","country_code":"ES","type":"education","lineage":["https://openalex.org/I63634437"]},{"id":"https://openalex.org/I88060688","display_name":"Universidad Polit\u00e9cnica de Madrid","ror":"https://ror.org/03n6nwv02","country_code":"ES","type":"education","lineage":["https://openalex.org/I88060688"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Alicia Lozano-Diez","raw_affiliation_strings":["AUDIAS, Electronic and Communication Technology Department, Escuela Polit\u00e9cnica Superior, Universidad Aut\u00f3noma de Madrid, Av. Francisco Tom\u00e1s y Valiente, 11, Madrid, 28049, Spain"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AUDIAS, Electronic and Communication Technology Department, Escuela Polit\u00e9cnica Superior, Universidad Aut\u00f3noma de Madrid, Av. Francisco Tom\u00e1s y Valiente, 11, Madrid, 28049, Spain","institution_ids":["https://openalex.org/I63634437","https://openalex.org/I88060688"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128126229","display_name":"Daniel Ramos","orcid":null},"institutions":[{"id":"https://openalex.org/I63634437","display_name":"Universidad Aut\u00f3noma de Madrid","ror":"https://ror.org/01cby8j38","country_code":"ES","type":"education","lineage":["https://openalex.org/I63634437"]},{"id":"https://openalex.org/I88060688","display_name":"Universidad Polit\u00e9cnica de Madrid","ror":"https://ror.org/03n6nwv02","country_code":"ES","type":"education","lineage":["https://openalex.org/I88060688"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Daniel Ramos","raw_affiliation_strings":["AUDIAS, Electronic and Communication Technology Department, Escuela Polit\u00e9cnica Superior, Universidad Aut\u00f3noma de Madrid, Av. Francisco Tom\u00e1s y Valiente, 11, Madrid, 28049, Spain"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AUDIAS, Electronic and Communication Technology Department, Escuela Polit\u00e9cnica Superior, Universidad Aut\u00f3noma de Madrid, Av. Francisco Tom\u00e1s y Valiente, 11, Madrid, 28049, Spain","institution_ids":["https://openalex.org/I63634437","https://openalex.org/I88060688"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089438058","display_name":"Doroteo T. Toledano","orcid":"https://orcid.org/0000-0003-1159-6455"},"institutions":[{"id":"https://openalex.org/I63634437","display_name":"Universidad Aut\u00f3noma de Madrid","ror":"https://ror.org/01cby8j38","country_code":"ES","type":"education","lineage":["https://openalex.org/I63634437"]},{"id":"https://openalex.org/I88060688","display_name":"Universidad Polit\u00e9cnica de Madrid","ror":"https://ror.org/03n6nwv02","country_code":"ES","type":"education","lineage":["https://openalex.org/I88060688"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Doroteo T. Toledano","raw_affiliation_strings":["AUDIAS, Electronic and Communication Technology Department, Escuela Polit\u00e9cnica Superior, Universidad Aut\u00f3noma de Madrid, Av. Francisco Tom\u00e1s y Valiente, 11, Madrid, 28049, Spain"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AUDIAS, Electronic and Communication Technology Department, Escuela Polit\u00e9cnica Superior, Universidad Aut\u00f3noma de Madrid, Av. Francisco Tom\u00e1s y Valiente, 11, Madrid, 28049, Spain","institution_ids":["https://openalex.org/I63634437","https://openalex.org/I88060688"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5106528898"],"corresponding_institution_ids":["https://openalex.org/I63634437","https://openalex.org/I88060688"],"apc_list":{"value":3400,"currency":"USD","value_usd":3400},"apc_paid":{"value":3400,"currency":"USD","value_usd":3400},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.23081791,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"100","issue":null,"first_page":"101967","last_page":"101967"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.6394000053405762,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.6394000053405762,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.025800000876188278,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T13083","display_name":"Advanced Text Analysis Techniques","score":0.02500000037252903,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/event","display_name":"Event (particle physics)","score":0.6074000000953674},{"id":"https://openalex.org/keywords/sound","display_name":"Sound (geography)","score":0.3262999951839447},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.28760001063346863},{"id":"https://openalex.org/keywords/detection-theory","display_name":"Detection theory","score":0.26010000705718994}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8621000051498413},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.6074000000953674},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4375},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3993000090122223},{"id":"https://openalex.org/C203718221","wikidata":"https://www.wikidata.org/wiki/Q491713","display_name":"Sound (geography)","level":2,"score":0.3262999951839447},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.28760001063346863},{"id":"https://openalex.org/C137270730","wikidata":"https://www.wikidata.org/wiki/Q120811","display_name":"Detection theory","level":3,"score":0.26010000705718994},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.24819999933242798},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.243599995970726},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.23389999568462372}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1016/j.csl.2026.101967","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.csl.2026.101967","pdf_url":null,"source":{"id":"https://openalex.org/S91252481","display_name":"Computer Speech & Language","issn_l":"0885-2308","issn":["0885-2308","1095-8363"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer Speech &amp; Language","raw_type":"journal-article"},{"id":"pmh:oai:repositorio.uam.es:10486/754400","is_oa":true,"landing_page_url":"https://hdl.handle.net/10486/754400","pdf_url":null,"source":{"id":"https://openalex.org/S4306400964","display_name":"Biblos-e Archivo (Universidad Aut\u00f3noma de Madrid)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I63634437","host_organization_name":"Universidad Aut\u00f3noma de Madrid","host_organization_lineage":["https://openalex.org/I63634437"],"host_organization_lineage_names":[],"type":"repository"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"research article"}],"best_oa_location":{"id":"doi:10.1016/j.csl.2026.101967","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.csl.2026.101967","pdf_url":null,"source":{"id":"https://openalex.org/S91252481","display_name":"Computer Speech & Language","issn_l":"0885-2308","issn":["0885-2308","1095-8363"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer Speech &amp; Language","raw_type":"journal-article"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.4218716025352478}],"awards":[],"funders":[{"id":"https://openalex.org/F4320315062","display_name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":11,"referenced_works":["https://openalex.org/W2566935005","https://openalex.org/W2593116425","https://openalex.org/W2771361008","https://openalex.org/W2959539607","https://openalex.org/W2964110616","https://openalex.org/W3015190346","https://openalex.org/W3141035251","https://openalex.org/W3197813307","https://openalex.org/W3203468141","https://openalex.org/W4221149441","https://openalex.org/W4224920041"],"related_works":[],"abstract_inverted_index":{"Sound":[0,138],"Event":[1],"Detection":[2,139],"(SED)":[3],"requires":[4],"models":[5,211],"that":[6,158],"can":[7,36,42,64],"accurately":[8],"localize":[9],"and":[10,47,147,167,176,231],"classify":[11],"overlapping":[12],"audio":[13],"events":[14],"within":[15,102],"complex":[16],"acoustic":[17],"environments.":[18],"Conformer-based":[19,225],"architectures":[20],"have":[21],"demonstrated":[22],"promising":[23,222],"performance":[24,131,207],"by":[25,218],"leveraging":[26],"self-attention":[27,74],"to":[28,194,209],"capture":[29],"long-range":[30,168],"dependencies.":[31],"However,":[32],"this":[33],"global":[34],"attention":[35,189],"be":[37],"accumulated":[38],"across":[39],"layers,":[40],"which":[41,89],"blur":[43],"local":[44],"temporal":[45,67,100,113,165],"boundaries":[46],"reduce":[48],"detection":[49],"accuracy,":[50],"especially":[51],"for":[52,224],"short":[53],"or":[54],"closely":[55],"spaced":[56],"events.":[57],"While":[58],"increasing":[59],"the":[60,69,85,93,99,123,135,153,159],"input":[61,94,109],"sequence":[62,95],"length":[63],"help":[65],"recover":[66],"detail,":[68],"quadratic":[70],"complexity":[71],"of":[72,199,229],"Conformers\u2019":[73],"significantly":[75],"increases":[76],"computational":[77,216],"costs.":[78],"To":[79],"address":[80],"this,":[81],"we":[82,186],"propose":[83],"integrating":[84],"Efficient":[86,160],"Conformer":[87,161,175],"architecture,":[88],"introduces":[90],"subsampling":[91],"along":[92],"length,":[96],"effectively":[97],"reducing":[98,215],"dimension":[101],"blocks.":[103],"This":[104],"design":[105],"enables":[106],"processing":[107],"longer":[108],"sequences":[110],"at":[111],"finer":[112],"resolution,":[114],"enhancing":[115],"localization":[116,144],"accuracy":[117],"without":[118],"extending":[119],"output":[120],"length.":[121],"Using":[122],"DCASE":[124],"Challenge":[125],"2023":[126],"Task":[127],"4":[128],"benchmark,":[129],"system":[130],"is":[132],"evaluated":[133],"via":[134],"threshold-independent":[136],"Polyphonic":[137],"Score":[140],"(PSDS),":[141],"measuring":[142],"both":[143],"precision":[145,230],"(PSDS1)":[146],"class":[148],"robustness":[149],"(PSDS2).":[150],"Experiments":[151],"on":[152],"DESED":[154],"validation":[155],"dataset":[156],"demonstrate":[157],"not":[162],"only":[163],"improves":[164],"resolution":[166],"dependency":[169],"modeling,":[170],"but":[171],"also":[172],"outperforms":[173],"standard":[174],"Convolutional":[177],"Recurrent":[178],"Neural":[179],"Network":[180],"(CRNN)":[181],"baselines":[182],"in":[183,227],"PSDS2.":[184],"Additionally,":[185],"explore":[187],"lightweight":[188],"mechanisms":[190],"employing":[191],"squeeze-and-excitation":[192],"blocks":[193],"emulate":[195],"frequency-axis":[196],"translation":[197],"invariance":[198],"Frequency":[200],"Dynamic":[201],"Convolutions":[202],"(FDY).":[203],"Our":[204],"approach":[205],"achieves":[206],"comparable":[208],"heavier":[210],"like":[212],"FDY+Conformer,":[213],"while":[214],"cost":[217],"over":[219],"69%,":[220],"showing":[221],"results":[223],"systems":[226],"terms":[228],"model":[232],"efficiency.":[233]},"counts_by_year":[],"updated_date":"2026-08-09T07:27:16.801131","created_date":"2026-03-05T00:00:00"}
