{"id":"https://openalex.org/W4408352361","doi":"https://doi.org/10.1109/icassp49660.2025.10890174","title":"Harnessing the Zero-Shot Power of Instruction-Tuned Large Language Model for Guiding End-to-End Speech Recognition","display_name":"Harnessing the Zero-Shot Power of Instruction-Tuned Large Language Model for Guiding End-to-End Speech Recognition","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408352361","doi":"https://doi.org/10.1109/icassp49660.2025.10890174"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10890174","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890174","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/A5102017467","display_name":"Yosuke Higuchi","orcid":"https://orcid.org/0000-0003-4500-8957"},"institutions":[{"id":"https://openalex.org/I150744194","display_name":"Waseda University","ror":"https://ror.org/00ntfnx83","country_code":"JP","type":"education","lineage":["https://openalex.org/I150744194"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Yosuke Higuchi","raw_affiliation_strings":["Waseda University,Department of Communications and Computer Engineering,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"Waseda University,Department of Communications and Computer Engineering,Tokyo,Japan","institution_ids":["https://openalex.org/I150744194"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087632404","display_name":"Tetsuji Ogawa","orcid":"https://orcid.org/0000-0002-7316-2073"},"institutions":[{"id":"https://openalex.org/I150744194","display_name":"Waseda University","ror":"https://ror.org/00ntfnx83","country_code":"JP","type":"education","lineage":["https://openalex.org/I150744194"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Tetsuji Ogawa","raw_affiliation_strings":["Waseda University,Department of Communications and Computer Engineering,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"Waseda University,Department of Communications and Computer Engineering,Tokyo,Japan","institution_ids":["https://openalex.org/I150744194"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101188700","display_name":"Tetsunori Kobayashi","orcid":null},"institutions":[{"id":"https://openalex.org/I150744194","display_name":"Waseda University","ror":"https://ror.org/00ntfnx83","country_code":"JP","type":"education","lineage":["https://openalex.org/I150744194"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Tetsunori Kobayashi","raw_affiliation_strings":["Waseda University,Department of Communications and Computer Engineering,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"Waseda University,Department of Communications and Computer Engineering,Tokyo,Japan","institution_ids":["https://openalex.org/I150744194"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5102017467"],"corresponding_institution_ids":["https://openalex.org/I150744194"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.02008316,"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":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9883999824523926,"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":0.9883999824523926,"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/end-to-end-principle","display_name":"End-to-end principle","score":0.891257643699646},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6792416572570801},{"id":"https://openalex.org/keywords/zero","display_name":"Zero (linguistics)","score":0.5873320698738098},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5245435833930969},{"id":"https://openalex.org/keywords/power","display_name":"Power (physics)","score":0.46482494473457336},{"id":"https://openalex.org/keywords/one-shot","display_name":"One shot","score":0.46167421340942383},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.44401049613952637},{"id":"https://openalex.org/keywords/end-of-history","display_name":"End of history","score":0.4268324673175812},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.27809250354766846},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.16933435201644897},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.1477639079093933},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1467326581478119}],"concepts":[{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.891257643699646},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6792416572570801},{"id":"https://openalex.org/C2780813799","wikidata":"https://www.wikidata.org/wiki/Q3274237","display_name":"Zero (linguistics)","level":2,"score":0.5873320698738098},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5245435833930969},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.46482494473457336},{"id":"https://openalex.org/C2992734406","wikidata":"https://www.wikidata.org/wiki/Q413267","display_name":"One shot","level":2,"score":0.46167421340942383},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.44401049613952637},{"id":"https://openalex.org/C2778935963","wikidata":"https://www.wikidata.org/wiki/Q13218530","display_name":"End of history","level":3,"score":0.4268324673175812},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.27809250354766846},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.16933435201644897},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.1477639079093933},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1467326581478119},{"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/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C78519656","wikidata":"https://www.wikidata.org/wiki/Q101333","display_name":"Mechanical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C94625758","wikidata":"https://www.wikidata.org/wiki/Q7163","display_name":"Politics","level":2,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10890174","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890174","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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":61,"referenced_works":["https://openalex.org/W179875071","https://openalex.org/W1494198834","https://openalex.org/W1985258458","https://openalex.org/W2127141656","https://openalex.org/W2327501763","https://openalex.org/W2526425061","https://openalex.org/W2739883972","https://openalex.org/W2766219058","https://openalex.org/W2888779557","https://openalex.org/W2939111082","https://openalex.org/W2939216642","https://openalex.org/W2962780374","https://openalex.org/W2963240019","https://openalex.org/W2973122799","https://openalex.org/W2979826702","https://openalex.org/W3011339933","https://openalex.org/W3024308166","https://openalex.org/W3097777922","https://openalex.org/W3163793923","https://openalex.org/W3196509775","https://openalex.org/W4224919446","https://openalex.org/W4297841327","https://openalex.org/W4297841367","https://openalex.org/W4372259777","https://openalex.org/W4385245566","https://openalex.org/W4385567350","https://openalex.org/W4385822949","https://openalex.org/W4389524500","https://openalex.org/W4391021505","https://openalex.org/W4391021666","https://openalex.org/W4391021684","https://openalex.org/W4391021773","https://openalex.org/W4392903956","https://openalex.org/W4392931626","https://openalex.org/W4402112323","https://openalex.org/W4404918643","https://openalex.org/W6617145748","https://openalex.org/W6623517193","https://openalex.org/W6640059789","https://openalex.org/W6640090968","https://openalex.org/W6675365184","https://openalex.org/W6691770337","https://openalex.org/W6743726175","https://openalex.org/W6755207826","https://openalex.org/W6778883912","https://openalex.org/W6800875267","https://openalex.org/W6810081322","https://openalex.org/W6810738896","https://openalex.org/W6838461927","https://openalex.org/W6847076894","https://openalex.org/W6847118041","https://openalex.org/W6851120679","https://openalex.org/W6851499912","https://openalex.org/W6852326057","https://openalex.org/W6852818750","https://openalex.org/W6853530687","https://openalex.org/W6853611000","https://openalex.org/W6853998256","https://openalex.org/W6854445763","https://openalex.org/W6854866820","https://openalex.org/W6860809563"],"related_works":["https://openalex.org/W3016188207","https://openalex.org/W3179968364","https://openalex.org/W4286908490","https://openalex.org/W4307058106","https://openalex.org/W2916997151","https://openalex.org/W2964709658","https://openalex.org/W2795393712","https://openalex.org/W4308474082","https://openalex.org/W4393212338","https://openalex.org/W4406318200"],"abstract_inverted_index":{"We":[0],"propose":[1],"to":[2,48,66,79,112,141],"utilize":[3],"an":[4,64,71],"instruction-tuned":[5],"large":[6],"language":[7],"model":[8,86,167],"(LLM)":[9],"for":[10,38,105],"guiding":[11],"the":[12,44,76,81,90,97,106,117,125,138,149,152,158,164],"text":[13,28,55],"generation":[14,29,56],"process":[15],"in":[16,57,70,175],"automatic":[17],"speech":[18],"recognition":[19],"(ASR).":[20],"Modern":[21],"LLMs":[22,47],"are":[23],"adept":[24],"at":[25],"performing":[26],"various":[27],"tasks":[30],"through":[31],"zero-shot":[32],"learning,":[33],"prompted":[34],"with":[35,96,128],"instructions":[36],"designed":[37],"specific":[39,130],"objectives.":[40],"This":[41],"paper":[42],"explores":[43],"potential":[45],"of":[46,172],"derive":[49],"linguistic":[50,154],"information":[51,147,155],"that":[52,163],"can":[53],"facilitate":[54],"end-to-end":[58],"ASR":[59,72,109],"models.":[60],"Specifically,":[61],"we":[62],"instruct":[63],"LLM":[65,98,126,139],"correct":[67],"grammatical":[68],"errors":[69],"hypothesis":[73],"and":[74,93,122,151],"use":[75],"LLM-derived":[77],"representations":[78],"refine":[80],"output":[82,140],"further.":[83],"The":[84,108,132],"proposed":[85,165],"is":[87,114],"built":[88],"on":[89],"joint":[91],"CTC":[92,120],"attention":[94],"architecture,":[95],"serving":[99],"as":[100,136],"a":[101,129,169],"front-end":[102],"feature":[103],"extractor":[104],"decoder.":[107],"hypothesis,":[110],"subject":[111],"correction,":[113],"obtained":[115],"from":[116,148],"encoder":[118,150],"via":[119],"decoding":[121],"fed":[123],"into":[124],"along":[127],"instruction.":[131],"decoder":[133],"subsequently":[134],"takes":[135],"input":[137],"perform":[142],"token":[143],"predictions,":[144],"combining":[145],"acoustic":[146],"powerful":[153],"provided":[156],"by":[157],"LLM.":[159],"Experimental":[160],"results":[161],"show":[162],"LLM-guided":[166],"achieves":[168],"relative":[170],"gain":[171],"approximately":[173],"13%":[174],"word":[176],"error":[177],"rates":[178],"across":[179],"major":[180],"benchmarks.":[181]},"counts_by_year":[],"updated_date":"2025-12-21T23:12:01.093139","created_date":"2025-10-10T00:00:00"}