{"id":"https://openalex.org/W4408354400","doi":"https://doi.org/10.1109/icassp49660.2025.10889878","title":"Efficient Finetuning for Dimensional Speech Emotion Recognition in the Age of Transformers","display_name":"Efficient Finetuning for Dimensional Speech Emotion Recognition in the Age of Transformers","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408354400","doi":"https://doi.org/10.1109/icassp49660.2025.10889878"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10889878","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889878","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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/A5085684309","display_name":"Aneesha Sampath","orcid":null},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan\u2013Ann Arbor","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Aneesha Sampath","raw_affiliation_strings":["University of Michigan,Computer Science and Engineering,Ann Arbor,USA"],"affiliations":[{"raw_affiliation_string":"University of Michigan,Computer Science and Engineering,Ann Arbor,USA","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092639639","display_name":"James Tavernor","orcid":"https://orcid.org/0000-0002-8100-231X"},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan\u2013Ann Arbor","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"James Tavernor","raw_affiliation_strings":["University of Michigan,Computer Science and Engineering,Ann Arbor,USA"],"affiliations":[{"raw_affiliation_string":"University of Michigan,Computer Science and Engineering,Ann Arbor,USA","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5003136334","display_name":"Emily Mower Provost","orcid":"https://orcid.org/0000-0003-1870-6063"},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan\u2013Ann Arbor","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Emily Mower Provost","raw_affiliation_strings":["University of Michigan,Computer Science and Engineering,Ann Arbor,USA"],"affiliations":[{"raw_affiliation_string":"University of Michigan,Computer Science and Engineering,Ann Arbor,USA","institution_ids":["https://openalex.org/I27837315"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5085684309"],"corresponding_institution_ids":["https://openalex.org/I27837315"],"apc_list":null,"apc_paid":null,"fwci":4.3172,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.93175624,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9699000120162964,"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":0.9699000120162964,"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/computer-science","display_name":"Computer science","score":0.6369350552558899},{"id":"https://openalex.org/keywords/emotion-recognition","display_name":"Emotion recognition","score":0.5828428268432617},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.49312660098075867},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4593821167945862},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.328888475894928},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08536046743392944},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.059549301862716675}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6369350552558899},{"id":"https://openalex.org/C2777438025","wikidata":"https://www.wikidata.org/wiki/Q1339090","display_name":"Emotion recognition","level":2,"score":0.5828428268432617},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.49312660098075867},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4593821167945862},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.328888475894928},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08536046743392944},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.059549301862716675},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10889878","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889878","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 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":26,"referenced_works":["https://openalex.org/W2040378199","https://openalex.org/W2239141610","https://openalex.org/W2313339984","https://openalex.org/W2742542661","https://openalex.org/W2747172199","https://openalex.org/W2757888873","https://openalex.org/W2979124240","https://openalex.org/W3197642003","https://openalex.org/W3200129129","https://openalex.org/W3202370288","https://openalex.org/W3206189675","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4226380987","https://openalex.org/W4297841899","https://openalex.org/W4361994820","https://openalex.org/W4375869379","https://openalex.org/W4390874621","https://openalex.org/W6745245109","https://openalex.org/W6757817989","https://openalex.org/W6762407294","https://openalex.org/W6780218876","https://openalex.org/W6787141514","https://openalex.org/W6796581206","https://openalex.org/W6803378298","https://openalex.org/W6869487681"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W3126677997","https://openalex.org/W1610857240"],"abstract_inverted_index":{"Accurate":[0],"speech":[1,181],"emotion":[2,42,62,182,187],"recognition":[3,43,183,188],"is":[4],"essential":[5],"for":[6,40,167],"developing":[7],"human-facing":[8],"systems.":[9],"Recent":[10],"advancements":[11],"have":[12],"included":[13],"finetuning":[14,24,48,80,86,90,95,112,121,147,180],"large,":[15],"pretrained":[16],"transformer":[17,88,151],"models":[18],"like":[19],"Wav2Vec":[20,104],"2.0.":[21],"However,":[22],"the":[23,38,54,103,148,177],"process":[25],"requires":[26],"substantial":[27],"computational":[28,55],"resources,":[29],"including":[30,82],"high-memory":[31],"GPUs":[32],"and":[33,72,98,138,159,162,197],"significant":[34],"processing":[35],"time.":[36],"As":[37],"demand":[39],"accurate":[41,186],"continues":[44],"to":[45,52,70,75,119,156,179,191],"grow,":[46],"efficient":[47],"approaches":[49],"are":[50],"needed":[51],"reduce":[53],"burden.":[56],"Our":[57],"study":[58],"focuses":[59],"on":[60,102],"dimensional":[61],"recognition,":[63],"predicting":[64],"attributes":[65],"such":[66],"as":[67],"activation":[68],"(calm":[69],"excited)":[71],"valence":[73],"(negative":[74],"positive).":[76],"We":[77,108,145],"present":[78],"various":[79],"techniques,":[81],"full":[83,120],"finetuning,":[84],"partial":[85,94,111],"of":[87,195],"layers,":[89],"with":[91,96,113,170],"mixed":[92,114,154],"precision,":[93],"caching,":[97],"low-rank":[99],"adaptation":[100],"(LoRA)":[101],"2.0":[105],"base":[106],"model.":[107],"find":[109],"that":[110],"precision":[115,155],"achieves":[116],"performance":[117,158,172],"comparable":[118],"while":[122],"increasing":[123],"training":[124,160],"speed":[125,169],"by":[126],"67%.":[127],"Caching":[128],"intermediate":[129,164],"representations":[130],"further":[131],"boosts":[132],"efficiency,":[133,161],"yielding":[134],"an":[135],"88%":[136],"speedup":[137],"a":[139,192],"71%":[140],"reduction":[141],"in":[142,153],"learnable":[143],"parameters.":[144],"recommend":[146],"final":[149],"three":[150],"layers":[152],"balance":[157],"adding":[163],"representation":[165],"caching":[166],"optimal":[168],"minimal":[171],"trade-offs.":[173],"These":[174],"findings":[175],"lower":[176],"barriers":[178],"systems,":[184],"making":[185],"more":[189],"accessible":[190],"broader":[193],"range":[194],"researchers":[196],"practitioners.":[198]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2}],"updated_date":"2026-04-03T22:45:19.894376","created_date":"2025-10-10T00:00:00"}