{"id":"https://openalex.org/W7148366777","doi":"https://doi.org/10.1109/asru65441.2025.11434659","title":"Unifying Model and Layer Fusion for Speech Foundation Models","display_name":"Unifying Model and Layer Fusion for Speech Foundation Models","publication_year":2025,"publication_date":"2025-12-06","ids":{"openalex":"https://openalex.org/W7148366777","doi":"https://doi.org/10.1109/asru65441.2025.11434659"},"language":null,"primary_location":{"id":"doi:10.1109/asru65441.2025.11434659","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru65441.2025.11434659","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","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/A5053466746","display_name":"Yi-Jen Shih","orcid":"https://orcid.org/0000-0003-3481-3117"},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Yi-Jen Shih","raw_affiliation_strings":["The University of Texas at Austin"],"affiliations":[{"raw_affiliation_string":"The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5004717608","display_name":"David Harwath","orcid":"https://orcid.org/0000-0003-0206-0253"},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David Harwath","raw_affiliation_strings":["The University of Texas at Austin"],"affiliations":[{"raw_affiliation_string":"The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5053466746"],"corresponding_institution_ids":["https://openalex.org/I86519309"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.87346039,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.7879999876022339,"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"}},"topics":[{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.7879999876022339,"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/T12031","display_name":"Speech and dialogue systems","score":0.05869999900460243,"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/T10028","display_name":"Topic Modeling","score":0.023600000888109207,"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/paralanguage","display_name":"Paralanguage","score":0.6894999742507935},{"id":"https://openalex.org/keywords/upstream","display_name":"Upstream (networking)","score":0.6488999724388123},{"id":"https://openalex.org/keywords/fusion","display_name":"Fusion","score":0.6051999926567078},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5464000105857849},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.4918999969959259},{"id":"https://openalex.org/keywords/interface","display_name":"Interface (matter)","score":0.4634000062942505},{"id":"https://openalex.org/keywords/downstream","display_name":"Downstream (manufacturing)","score":0.38199999928474426}],"concepts":[{"id":"https://openalex.org/C133378560","wikidata":"https://www.wikidata.org/wiki/Q1753225","display_name":"Paralanguage","level":2,"score":0.6894999742507935},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6765000224113464},{"id":"https://openalex.org/C191172861","wikidata":"https://www.wikidata.org/wiki/Q7899321","display_name":"Upstream (networking)","level":2,"score":0.6488999724388123},{"id":"https://openalex.org/C158525013","wikidata":"https://www.wikidata.org/wiki/Q2593739","display_name":"Fusion","level":2,"score":0.6051999926567078},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5464000105857849},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.4918999969959259},{"id":"https://openalex.org/C113843644","wikidata":"https://www.wikidata.org/wiki/Q901882","display_name":"Interface (matter)","level":4,"score":0.4634000062942505},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45579999685287476},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.38199999928474426},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3779999911785126},{"id":"https://openalex.org/C33954974","wikidata":"https://www.wikidata.org/wiki/Q486494","display_name":"Sensor fusion","level":2,"score":0.37450000643730164},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.3725000023841858},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.3644999861717224},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33390000462532043},{"id":"https://openalex.org/C45804977","wikidata":"https://www.wikidata.org/wiki/Q7239673","display_name":"Predictive modelling","level":2,"score":0.319599986076355},{"id":"https://openalex.org/C14999030","wikidata":"https://www.wikidata.org/wiki/Q16346","display_name":"Speech synthesis","level":2,"score":0.30140000581741333},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.25380000472068787},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.2515000104904175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asru65441.2025.11434659","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru65441.2025.11434659","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.47201281785964966,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":23,"referenced_works":["https://openalex.org/W2093717447","https://openalex.org/W2135293965","https://openalex.org/W2912934387","https://openalex.org/W3197580070","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4223503417","https://openalex.org/W4224875474","https://openalex.org/W4225635674","https://openalex.org/W4297841583","https://openalex.org/W4319862440","https://openalex.org/W4319862477","https://openalex.org/W4319862670","https://openalex.org/W4372267276","https://openalex.org/W4375869259","https://openalex.org/W4385822439","https://openalex.org/W4385822574","https://openalex.org/W4385823172","https://openalex.org/W4394862910","https://openalex.org/W4402111453","https://openalex.org/W4402111598","https://openalex.org/W4402112390","https://openalex.org/W4402112493"],"related_works":[],"abstract_inverted_index":{"Speech":[0,133],"Foundation":[1,134],"Models":[2],"have":[3,10],"gained":[4],"significant":[5],"attention":[6],"recently.":[7],"Prior":[8],"works":[9],"shown":[11],"that":[12,47,82,110],"the":[13,21,25,100,111],"fusion":[14,26,40,49,87],"of":[15,20,27,102],"representations":[16],"from":[17],"multiple":[18,28,51],"layers":[19],"same":[22],"model":[23,95,124],"or":[24],"models":[29,54,70],"can":[30],"improve":[31],"performance":[32,117],"on":[33,65],"downstream":[34],"tasks.":[35],"We":[36,61,89],"unify":[37],"these":[38],"two":[39],"strategies":[41],"by":[42],"proposing":[43],"an":[44,115],"interface":[45,113],"module":[46],"enables":[48],"across":[50,58,71],"upstream":[52,105,123],"speech":[53,73],"while":[55],"integrating":[56],"information":[57],"their":[59],"layers.":[60],"conduct":[62],"extensive":[63],"experiments":[64],"different":[66],"self-supervised":[67],"and":[68,77,80,97],"supervised":[69],"various":[72],"tasks,":[74],"including":[75],"ASR":[76],"paralinguistic":[78],"analysis,":[79],"demonstrate":[81],"our":[83],"method":[84],"outperforms":[85],"prior":[86],"approaches.":[88],"further":[90],"analyze":[91],"its":[92],"scalability":[93],"concerning":[94],"size":[96],"count,":[98],"highlighting":[99],"importance":[101],"selecting":[103],"appropriate":[104],"models.":[106],"Our":[107],"results":[108],"show":[109],"proposed":[112],"provides":[114],"additional":[116],"boost":[118],"when":[119],"given":[120],"a":[121,128],"suitable":[122],"selection,":[125],"making":[126],"it":[127],"promising":[129],"approach":[130],"for":[131],"utilizing":[132],"Models.":[135]},"counts_by_year":[],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2026-04-03T00:00:00"}