{"id":"https://openalex.org/W4392904276","doi":"https://doi.org/10.1109/icassp48485.2024.10447390","title":"AttHear: Explaining Audio Transformers Using Attention-Aware NMF","display_name":"AttHear: Explaining Audio Transformers Using Attention-Aware NMF","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392904276","doi":"https://doi.org/10.1109/icassp48485.2024.10447390"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10447390","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447390","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069525898","display_name":"Alican Akman","orcid":"https://orcid.org/0000-0002-8010-6897"},"institutions":[{"id":"https://openalex.org/I47508984","display_name":"Imperial College London","ror":"https://ror.org/041kmwe10","country_code":"GB","type":"education","lineage":["https://openalex.org/I47508984"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Alican Akman","raw_affiliation_strings":["Imperial College London,GLAM,Department of Computing,London,UK","Department of Computing, GLAM, Imperial College London, London, UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Imperial College London,GLAM,Department of Computing,London,UK","institution_ids":["https://openalex.org/I47508984"]},{"raw_affiliation_string":"Department of Computing, GLAM, Imperial College London, London, UK","institution_ids":["https://openalex.org/I47508984"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5043060302","display_name":"Bj\u00f6rn W. Schuller","orcid":"https://orcid.org/0000-0002-6478-8699"},"institutions":[{"id":"https://openalex.org/I47508984","display_name":"Imperial College London","ror":"https://ror.org/041kmwe10","country_code":"GB","type":"education","lineage":["https://openalex.org/I47508984"]},{"id":"https://openalex.org/I62916508","display_name":"Technical University of Munich","ror":"https://ror.org/02kkvpp62","country_code":"DE","type":"education","lineage":["https://openalex.org/I62916508"]}],"countries":["DE","GB"],"is_corresponding":false,"raw_author_name":"Bj\u00f6rn W. Schuller","raw_affiliation_strings":["Imperial College London,GLAM,Department of Computing,London,UK","Chair of Health Informatics, MRI, Technical University of Munich, Germany","Department of Computing, GLAM, Imperial College London, London, UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Imperial College London,GLAM,Department of Computing,London,UK","institution_ids":["https://openalex.org/I47508984"]},{"raw_affiliation_string":"Chair of Health Informatics, MRI, Technical University of Munich, Germany","institution_ids":["https://openalex.org/I62916508"]},{"raw_affiliation_string":"Department of Computing, GLAM, Imperial College London, London, UK","institution_ids":["https://openalex.org/I47508984"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7015","last_page":"7019"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9994999766349792,"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":0.9994999766349792,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9743000268936157,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11439","display_name":"Video Analysis and Summarization","score":0.973800003528595,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/non-negative-matrix-factorization","display_name":"Non-negative matrix factorization","score":0.8077845573425293},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7930527329444885},{"id":"https://openalex.org/keywords/keyword-spotting","display_name":"Keyword spotting","score":0.6698998808860779},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6066927909851074},{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.5009901523590088},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.49404388666152954},{"id":"https://openalex.org/keywords/spotting","display_name":"Spotting","score":0.48818331956863403},{"id":"https://openalex.org/keywords/audio-signal-processing","display_name":"Audio signal processing","score":0.4593513309955597},{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.4522058963775635},{"id":"https://openalex.org/keywords/audio-visual","display_name":"Audio visual","score":0.4317355751991272},{"id":"https://openalex.org/keywords/matrix-decomposition","display_name":"Matrix decomposition","score":0.4117845296859741},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4068711996078491},{"id":"https://openalex.org/keywords/speech-coding","display_name":"Speech coding","score":0.3640941381454468},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.342521071434021},{"id":"https://openalex.org/keywords/audio-signal","display_name":"Audio signal","score":0.32666337490081787},{"id":"https://openalex.org/keywords/multimedia","display_name":"Multimedia","score":0.22721251845359802},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08817708492279053}],"concepts":[{"id":"https://openalex.org/C152671427","wikidata":"https://www.wikidata.org/wiki/Q10843505","display_name":"Non-negative matrix factorization","level":4,"score":0.8077845573425293},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7930527329444885},{"id":"https://openalex.org/C2781213101","wikidata":"https://www.wikidata.org/wiki/Q6398558","display_name":"Keyword spotting","level":2,"score":0.6698998808860779},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6066927909851074},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.5009901523590088},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.49404388666152954},{"id":"https://openalex.org/C2779506182","wikidata":"https://www.wikidata.org/wiki/Q7580141","display_name":"Spotting","level":2,"score":0.48818331956863403},{"id":"https://openalex.org/C127220857","wikidata":"https://www.wikidata.org/wiki/Q2719318","display_name":"Audio signal processing","level":4,"score":0.4593513309955597},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.4522058963775635},{"id":"https://openalex.org/C3017588708","wikidata":"https://www.wikidata.org/wiki/Q758901","display_name":"Audio visual","level":2,"score":0.4317355751991272},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.4117845296859741},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4068711996078491},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.3640941381454468},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.342521071434021},{"id":"https://openalex.org/C64922751","wikidata":"https://www.wikidata.org/wiki/Q4650799","display_name":"Audio signal","level":3,"score":0.32666337490081787},{"id":"https://openalex.org/C49774154","wikidata":"https://www.wikidata.org/wiki/Q131765","display_name":"Multimedia","level":1,"score":0.22721251845359802},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08817708492279053},{"id":"https://openalex.org/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.0},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.0},{"id":"https://openalex.org/C162853370","wikidata":"https://www.wikidata.org/wiki/Q39809","display_name":"Marketing","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C36289849","wikidata":"https://www.wikidata.org/wiki/Q34749","display_name":"Social science","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10447390","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447390","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":31,"referenced_works":["https://openalex.org/W1787224781","https://openalex.org/W2052666245","https://openalex.org/W2135029798","https://openalex.org/W2336525064","https://openalex.org/W2797583228","https://openalex.org/W2872172641","https://openalex.org/W2907159568","https://openalex.org/W2950784811","https://openalex.org/W2979476256","https://openalex.org/W3035422918","https://openalex.org/W3036601975","https://openalex.org/W3148040514","https://openalex.org/W3196974791","https://openalex.org/W3206494219","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4220714107","https://openalex.org/W4294975502","https://openalex.org/W4306955484","https://openalex.org/W4367365913","https://openalex.org/W4372266987","https://openalex.org/W4387966251","https://openalex.org/W4391486587","https://openalex.org/W6680012447","https://openalex.org/W6750665317","https://openalex.org/W6753441104","https://openalex.org/W6769196770","https://openalex.org/W6780218876","https://openalex.org/W6781206141","https://openalex.org/W6803164887","https://openalex.org/W6852010021"],"related_works":["https://openalex.org/W2918559346","https://openalex.org/W2114097550","https://openalex.org/W4286904253","https://openalex.org/W3119978414","https://openalex.org/W2516975559","https://openalex.org/W3206647229","https://openalex.org/W1969408022","https://openalex.org/W2000885660","https://openalex.org/W2545741539","https://openalex.org/W2172909489"],"abstract_inverted_index":{"The":[0,121],"increasing":[1],"success":[2],"of":[3],"transformer":[4,37,79],"models":[5,38,90],"in":[6,49],"various":[7],"fields,":[8],"such":[9],"as":[10],"computer":[11],"vision":[12],"and":[13,44,64,152],"audio":[14,53,100,105,126,137],"processing,":[15],"has":[16],"led":[17],"to":[18,25,58,76,96,113],"a":[19,73],"growing":[20],"need":[21],"for":[22,35,99,117,128],"improved":[23],"explainability":[24],"understand":[26],"their":[27,59],"complex":[28],"decision-making":[29],"processes":[30],"better.":[31],"Most":[32],"existing":[33],"techniques":[34],"explaining":[36],"concentrate":[39],"primarily":[40],"on":[41,61],"delivering":[42],"visual":[43,50],"textual":[45],"explanations,":[46],"commonly":[47],"used":[48],"media.":[51],"However,":[52],"explanations":[54,127,143],"are":[55,111],"crucial":[56],"due":[57],"intuitiveness":[60],"audio-based":[62],"tasks":[63],"distinguishing":[65],"expressiveness":[66],"over":[67],"other":[68],"modalities.":[69],"This":[70],"work":[71],"proposes":[72],"novel":[74],"method":[75,82,122],"interpret":[77],"audio-processing":[78],"models.":[80],"Our":[81,139],"combines":[83],"the":[84,129,134],"available":[85],"attention":[86,109],"mechanism":[87],"inside":[88],"these":[89,118],"with":[91],"non-negative":[92],"matrix":[93],"factorisation":[94],"(NMF)":[95],"compute":[97],"relevancy":[98],"inputs.":[101],"While":[102],"NMF":[103],"decomposes":[104],"into":[106],"spectral":[107,119],"patterns,":[108],"weights":[110],"utilised":[112],"calculate":[114],"time":[115],"activation":[116],"patterns.":[120],"then":[123],"generates":[124,142],"listenable":[125],"model\u2019s":[130],"final":[131],"decision":[132],"using":[133],"most":[135],"relevant":[136],"portions.":[138],"model":[140],"effectively":[141],"by":[144],"benchmarking":[145],"against":[146],"standard":[147],"datasets,":[148],"including":[149],"keyword":[150],"spotting":[151],"environmental":[153],"sound":[154],"classification.":[155]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2025-10-10T00:00:00"}