{"id":"https://openalex.org/W4404037515","doi":"https://doi.org/10.1109/mlsp58920.2024.10734776","title":"Parameter-Efficient Transfer Learning of Audio Spectrogram Transformers","display_name":"Parameter-Efficient Transfer Learning of Audio Spectrogram Transformers","publication_year":2024,"publication_date":"2024-09-22","ids":{"openalex":"https://openalex.org/W4404037515","doi":"https://doi.org/10.1109/mlsp58920.2024.10734776"},"language":"en","primary_location":{"id":"doi:10.1109/mlsp58920.2024.10734776","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mlsp58920.2024.10734776","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 34th International Workshop on Machine Learning for Signal Processing (MLSP)","raw_type":"proceedings-article"},"type":"article","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/A5043797502","display_name":"Umberto Cappellazzo","orcid":"https://orcid.org/0009-0006-3443-5143"},"institutions":[{"id":"https://openalex.org/I193223587","display_name":"University of Trento","ror":"https://ror.org/05trd4x28","country_code":"IT","type":"education","lineage":["https://openalex.org/I193223587"]}],"countries":["IT"],"is_corresponding":true,"raw_author_name":"Umberto Cappellazzo","raw_affiliation_strings":["University of Trento,Italy"],"affiliations":[{"raw_affiliation_string":"University of Trento,Italy","institution_ids":["https://openalex.org/I193223587"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041102462","display_name":"Daniele Falavigna","orcid":"https://orcid.org/0000-0002-4844-5071"},"institutions":[{"id":"https://openalex.org/I2277624104","display_name":"Fondazione Bruno Kessler","ror":"https://ror.org/01j33xk10","country_code":"IT","type":"facility","lineage":["https://openalex.org/I2277624104"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Daniele Falavigna","raw_affiliation_strings":["Fondazione Bruno Kessler,Trento"],"affiliations":[{"raw_affiliation_string":"Fondazione Bruno Kessler,Trento","institution_ids":["https://openalex.org/I2277624104"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066363315","display_name":"Alessio Brutti","orcid":"https://orcid.org/0000-0003-4146-3071"},"institutions":[{"id":"https://openalex.org/I2277624104","display_name":"Fondazione Bruno Kessler","ror":"https://ror.org/01j33xk10","country_code":"IT","type":"facility","lineage":["https://openalex.org/I2277624104"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Alessio Brutti","raw_affiliation_strings":["Fondazione Bruno Kessler,Trento"],"affiliations":[{"raw_affiliation_string":"Fondazione Bruno Kessler,Trento","institution_ids":["https://openalex.org/I2277624104"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5040811098","display_name":"Mirco Ravanelli","orcid":"https://orcid.org/0000-0002-3929-5526"},"institutions":[{"id":"https://openalex.org/I60158472","display_name":"Concordia University","ror":"https://ror.org/0420zvk78","country_code":"CA","type":"education","lineage":["https://openalex.org/I60158472"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Mirco Ravanelli","raw_affiliation_strings":["Concordia University,Canada"],"affiliations":[{"raw_affiliation_string":"Concordia University,Canada","institution_ids":["https://openalex.org/I60158472"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5043797502"],"corresponding_institution_ids":["https://openalex.org/I193223587"],"apc_list":null,"apc_paid":null,"fwci":2.616,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.91181667,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9886000156402588,"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/T10860","display_name":"Speech and Audio Processing","score":0.9886000156402588,"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/T12537","display_name":"Flow Measurement and Analysis","score":0.9258999824523926,"subfield":{"id":"https://openalex.org/subfields/2211","display_name":"Mechanics of Materials"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9182999730110168,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/spectrogram","display_name":"Spectrogram","score":0.9131587743759155},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6992887258529663},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5503649711608887},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4715828001499176},{"id":"https://openalex.org/keywords/transfer-of-learning","display_name":"Transfer of learning","score":0.45451048016548157},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3007339835166931},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.162395179271698},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1528606116771698},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.060264915227890015}],"concepts":[{"id":"https://openalex.org/C45273575","wikidata":"https://www.wikidata.org/wiki/Q578970","display_name":"Spectrogram","level":2,"score":0.9131587743759155},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6992887258529663},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5503649711608887},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4715828001499176},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.45451048016548157},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3007339835166931},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.162395179271698},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1528606116771698},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.060264915227890015}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/mlsp58920.2024.10734776","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mlsp58920.2024.10734776","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 34th International Workshop on Machine Learning for Signal Processing (MLSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W2038484192","https://openalex.org/W2052666245","https://openalex.org/W2146334809","https://openalex.org/W2797583228","https://openalex.org/W2972584841","https://openalex.org/W3097777922","https://openalex.org/W3153675281","https://openalex.org/W3173788106","https://openalex.org/W3174770825","https://openalex.org/W3176828726","https://openalex.org/W3196974791","https://openalex.org/W3201225328","https://openalex.org/W3206996142","https://openalex.org/W4205991051","https://openalex.org/W4224930323","https://openalex.org/W4285595687","https://openalex.org/W4312651322","https://openalex.org/W4319862642","https://openalex.org/W4361230739","https://openalex.org/W4372346804","https://openalex.org/W4385570973","https://openalex.org/W4385822750","https://openalex.org/W4385823383","https://openalex.org/W4390047809","https://openalex.org/W4401597540","https://openalex.org/W4402671897","https://openalex.org/W4402683929","https://openalex.org/W6750665317","https://openalex.org/W6759579507","https://openalex.org/W6780218876","https://openalex.org/W6780226713","https://openalex.org/W6795933390","https://openalex.org/W6796581206","https://openalex.org/W6797927333","https://openalex.org/W6802744804","https://openalex.org/W6838701581","https://openalex.org/W6839670772","https://openalex.org/W6851629679","https://openalex.org/W6859529408","https://openalex.org/W6861489518"],"related_works":["https://openalex.org/W2530685530","https://openalex.org/W4375868962","https://openalex.org/W2011227383","https://openalex.org/W2088854863","https://openalex.org/W4402568167","https://openalex.org/W3179495260","https://openalex.org/W1976719989","https://openalex.org/W2897924318","https://openalex.org/W2138997758","https://openalex.org/W3123837699"],"abstract_inverted_index":{"Parameter-efficient":[0],"transfer":[1,161],"learning":[2],"(PETL)":[3],"methods":[4,91],"have":[5],"emerged":[6],"as":[7],"a":[8,20,38,85,105],"solid":[9],"alternative":[10],"to":[11,63,67,99,119,155,180],"the":[12,34,41,51,93,96,111,115,123,141,169,172],"standard":[13,124],"full":[14,134],"fine-tuning":[15,135],"approach.":[16],"They":[17],"only":[18,138],"train":[19],"few":[21],"extra":[22],"parameters":[23],"for":[24,44,92],"each":[25,45],"downstream":[26,69],"task,":[27],"without":[28],"sacrificing":[29],"performance":[30,121,131],"and":[31,83,127,175],"dispensing":[32],"with":[33,133],"issue":[35],"of":[36,40,88,95,114,140,168,171],"storing":[37],"copy":[39],"pre-trained":[42,182],"model":[43,56,98],"task.":[46],"For":[47],"audio":[48],"classification":[49],"tasks,":[50],"Audio":[52],"Spectrogram":[53],"Transformer":[54],"(AST)":[55],"shows":[57],"impressive":[58],"results.":[59],"However,":[60],"surprisingly,":[61],"how":[62],"efficiently":[64],"adapt":[65],"it":[66],"several":[68],"tasks":[70],"has":[71],"not":[72],"been":[73],"tackled":[74],"before.":[75],"In":[76],"this":[77,81],"paper,":[78],"we":[79,103,144],"bridge":[80],"gap":[82],"present":[84],"detailed":[86],"investigation":[87],"common":[89],"PETL":[90,125],"adaptation":[94],"AST":[97],"audio/speech":[100],"tasks.":[101],"Furthermore,":[102],"propose":[104],"new":[106],"adapter":[107],"design":[108],"that":[109,149],"exploits":[110],"convolution":[112],"module":[113],"Conformer":[116],"model,":[117],"leading":[118],"superior":[120],"over":[122],"approaches":[126],"surpassing":[128],"or":[129],"achieving":[130],"parity":[132],"by":[136],"updating":[137],"0.29%":[139],"parameters.":[142],"Finally,":[143],"provide":[145],"ablation":[146],"studies":[147],"revealing":[148],"our":[150],"proposed":[151],"adapter:":[152],"1)":[153],"proves":[154],"be":[156,178],"effective":[157],"in":[158],"few-shot":[159],"efficient":[160],"learning,":[162],"2)":[163],"attains":[164],"optimal":[165],"results":[166],"regardless":[167],"amount":[170],"allocated":[173],"parameters,":[174],"3)":[176],"can":[177],"applied":[179],"other":[181],"models.":[183],"Our":[184],"code":[185],"is":[186],"available":[187],"at":[188],"https:/github.com/umbertocappellazzo/PETL_AST.":[189]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":6}],"updated_date":"2026-02-25T23:00:34.991745","created_date":"2025-10-10T00:00:00"}