{"id":"https://openalex.org/W4416799322","doi":"https://doi.org/10.1109/apsipaasc65261.2025.11249087","title":"HIPA-MoE: A Parameter-Efficient Fine-Tuning Architecture with Hierarchical Adapter-Based Mixture-Of-Experts for Multilingual ASR","display_name":"HIPA-MoE: A Parameter-Efficient Fine-Tuning Architecture with Hierarchical Adapter-Based Mixture-Of-Experts for Multilingual ASR","publication_year":2025,"publication_date":"2025-10-22","ids":{"openalex":"https://openalex.org/W4416799322","doi":"https://doi.org/10.1109/apsipaasc65261.2025.11249087"},"language":null,"primary_location":{"id":"doi:10.1109/apsipaasc65261.2025.11249087","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc65261.2025.11249087","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","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/A5111739631","display_name":"Xun Lu","orcid":"https://orcid.org/0000-0002-4992-0241"},"institutions":[{"id":"https://openalex.org/I4210099069","display_name":"Institute of Acoustics","ror":"https://ror.org/00v8rqv75","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210099069"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xun Lu","raw_affiliation_strings":["Institute of Acoustics, CAS,Key Laboratory of Speech Acoustics and Content Understanding,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Acoustics, CAS,Key Laboratory of Speech Acoustics and Content Understanding,China","institution_ids":["https://openalex.org/I4210099069"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100451347","display_name":"Xuyang Wang","orcid":"https://orcid.org/0009-0009-0061-4233"},"institutions":[{"id":"https://openalex.org/I4210099069","display_name":"Institute of Acoustics","ror":"https://ror.org/00v8rqv75","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210099069"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuyang Wang","raw_affiliation_strings":["Institute of Acoustics, CAS,Key Laboratory of Speech Acoustics and Content Understanding,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Acoustics, CAS,Key Laboratory of Speech Acoustics and Content Understanding,China","institution_ids":["https://openalex.org/I4210099069"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072887461","display_name":"Gaofeng Cheng","orcid":"https://orcid.org/0000-0002-2102-6061"},"institutions":[{"id":"https://openalex.org/I4210099069","display_name":"Institute of Acoustics","ror":"https://ror.org/00v8rqv75","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210099069"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Gaofeng Cheng","raw_affiliation_strings":["Institute of Acoustics, CAS,Key Laboratory of Speech Acoustics and Content Understanding,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Acoustics, CAS,Key Laboratory of Speech Acoustics and Content Understanding,China","institution_ids":["https://openalex.org/I4210099069"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028524428","display_name":"Lin Zheng","orcid":"https://orcid.org/0000-0001-8820-8145"},"institutions":[{"id":"https://openalex.org/I4210099069","display_name":"Institute of Acoustics","ror":"https://ror.org/00v8rqv75","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210099069"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lin Zheng","raw_affiliation_strings":["Institute of Acoustics, CAS,Key Laboratory of Speech Acoustics and Content Understanding,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Acoustics, CAS,Key Laboratory of Speech Acoustics and Content Understanding,China","institution_ids":["https://openalex.org/I4210099069"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5036446253","display_name":"Pengyuan Zhang","orcid":"https://orcid.org/0000-0001-6838-5160"},"institutions":[{"id":"https://openalex.org/I4210099069","display_name":"Institute of Acoustics","ror":"https://ror.org/00v8rqv75","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210099069"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Pengyuan Zhang","raw_affiliation_strings":["Institute of Acoustics, CAS,Key Laboratory of Speech Acoustics and Content Understanding,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Acoustics, CAS,Key Laboratory of Speech Acoustics and Content Understanding,China","institution_ids":["https://openalex.org/I4210099069"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210099069"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.17294126,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1068","last_page":"1073"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9641000032424927,"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":0.9641000032424927,"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/T10860","display_name":"Speech and Audio Processing","score":0.009999999776482582,"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/T10403","display_name":"Phonetics and Phonology Research","score":0.0019000000320374966,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/adapter","display_name":"Adapter (computing)","score":0.8281000256538391},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.5468000173568726},{"id":"https://openalex.org/keywords/feed-forward","display_name":"Feed forward","score":0.3747999966144562},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.35760000348091125},{"id":"https://openalex.org/keywords/pyramid","display_name":"Pyramid (geometry)","score":0.3068999946117401},{"id":"https://openalex.org/keywords/voice-activity-detection","display_name":"Voice activity detection","score":0.3012999892234802}],"concepts":[{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.8281000256538391},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8273000121116638},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6251000165939331},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.5468000173568726},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3937000036239624},{"id":"https://openalex.org/C38858127","wikidata":"https://www.wikidata.org/wiki/Q5441228","display_name":"Feed forward","level":2,"score":0.3747999966144562},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.35760000348091125},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.34470000863075256},{"id":"https://openalex.org/C142575187","wikidata":"https://www.wikidata.org/wiki/Q3358290","display_name":"Pyramid (geometry)","level":2,"score":0.3068999946117401},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.3012999892234802},{"id":"https://openalex.org/C2780844864","wikidata":"https://www.wikidata.org/wiki/Q184377","display_name":"Pronunciation","level":2,"score":0.2906999886035919},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.27489998936653137},{"id":"https://openalex.org/C18555067","wikidata":"https://www.wikidata.org/wiki/Q8375051","display_name":"Joint (building)","level":2,"score":0.2703999876976013},{"id":"https://openalex.org/C98025372","wikidata":"https://www.wikidata.org/wiki/Q477538","display_name":"Systems architecture","level":3,"score":0.26840001344680786},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.2630999982357025},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.25189998745918274}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/apsipaasc65261.2025.11249087","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc65261.2025.11249087","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3114329710","display_name":null,"funder_award_id":"62401560","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8845216468","display_name":null,"funder_award_id":"2024YFF0907304","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W2804732445","https://openalex.org/W2933138175","https://openalex.org/W4385822875","https://openalex.org/W4385823078","https://openalex.org/W4392902792","https://openalex.org/W4392903557","https://openalex.org/W4392904112","https://openalex.org/W4402115988","https://openalex.org/W4404782746","https://openalex.org/W4408345589","https://openalex.org/W4408345957","https://openalex.org/W4408347292","https://openalex.org/W4408352399","https://openalex.org/W4408353886","https://openalex.org/W4408354530","https://openalex.org/W4409362710"],"related_works":[],"abstract_inverted_index":{"Multilingual":[0],"automatic":[1,33],"speech":[2,34,121],"recognition":[3],"(MASR)":[4],"has":[5,26],"advanced":[6],"significantly":[7],"with":[8,70],"self-supervised":[9],"pretraining":[10],"(SSL).":[11],"However,":[12],"conventional":[13],"fine-tuning":[14],"remains":[15],"constrained":[16],"by":[17],"data":[18],"imbalance,":[19],"especially":[20],"for":[21,31,105],"low-resource":[22],"languages.":[23],"Although":[24],"Mixture-of-Experts(MoE)":[25],"provided":[27],"a":[28,55],"promising":[29],"approach":[30],"multilingual":[32],"recognition,":[35],"expanding":[36],"feedforward":[37],"networks":[38],"(FFNs)":[39],"into":[40],"MoE":[41],"layers":[42,94,101],"often":[43],"incurs":[44],"significant":[45],"parameter":[46,141],"overhead.":[47],"To":[48],"solve":[49],"these":[50],"issues,":[51],"we":[52,109],"propose":[53],"HIPAMoE,":[54],"novel":[56],"Hierarchical":[57],"Inverted":[58],"Pyramid":[59],"Adapter":[60],"Mixture-ofExperts":[61],"(HIPA-MoE)":[62],"architecture":[63],"that":[64,128],"uses":[65],"lightweight":[66],"adapters":[67,90,104],"as":[68],"experts":[69,77],"language-aware":[71],"routing.":[72],"Our":[73],"model":[74,95],"hierarchically":[75],"organizes":[76],"to":[78,118],"capture":[79],"both":[80],"universal":[81],"acoustic":[82],"features":[83],"and":[84,135,143],"language-specific":[85,103],"phonetic":[86],"nuances.":[87],"Specifically,":[88],"shared":[89],"in":[91,133],"the":[92,99],"lower":[93],"cross-lingual":[96,120],"patterns,":[97],"while":[98,138],"upper":[100],"deploy":[102],"fine-grained":[106],"specialization.":[107],"Additionally,":[108],"incorporate":[110],"an":[111],"auxiliary":[112],"phoneme-level":[113],"loss":[114],"via":[115],"uroman":[116],"transliterations":[117],"enhance":[119],"representation.":[122],"Experiments":[123],"on":[124],"ML-SUPERB":[125],"2.0":[126],"demonstrate":[127],"HIPA-MoE":[129],"achieves":[130],"state-of-the-art":[131],"performance":[132],"lowresource":[134],"long-tail":[136],"languages":[137],"maintaining":[139],"high":[140],"efficiency":[142],"scalability.":[144]},"counts_by_year":[],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-11-28T00:00:00"}