{"id":"https://openalex.org/W4408353954","doi":"https://doi.org/10.1109/icassp49660.2025.10888566","title":"LLM based Text Generation for Improved Low-resource Speech Recognition Models","display_name":"LLM based Text Generation for Improved Low-resource Speech Recognition Models","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408353954","doi":"https://doi.org/10.1109/icassp49660.2025.10888566"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10888566","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888566","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/A5069028016","display_name":"Tohru Nagano","orcid":"https://orcid.org/0000-0002-3686-2791"},"institutions":[{"id":"https://openalex.org/I4210145865","display_name":"IBM Research - Tokyo","ror":"https://ror.org/04915qk43","country_code":"JP","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115","https://openalex.org/I4210145865"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Tohru Nagano","raw_affiliation_strings":["IBM Research,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"IBM Research,Tokyo,Japan","institution_ids":["https://openalex.org/I4210145865"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021294422","display_name":"Gakuto Kurata","orcid":null},"institutions":[{"id":"https://openalex.org/I4210145865","display_name":"IBM Research - Tokyo","ror":"https://ror.org/04915qk43","country_code":"JP","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115","https://openalex.org/I4210145865"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Gakuto Kurata","raw_affiliation_strings":["IBM Research,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"IBM Research,Tokyo,Japan","institution_ids":["https://openalex.org/I4210145865"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101787514","display_name":"Samuel Thomas","orcid":"https://orcid.org/0000-0001-7573-0620"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Samuel Thomas","raw_affiliation_strings":["IBM Research,USA"],"affiliations":[{"raw_affiliation_string":"IBM Research,USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110252428","display_name":"Hong-Kwang Jeff Kuo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hong-Kwang J. Kuo","raw_affiliation_strings":["IBM Research,USA"],"affiliations":[{"raw_affiliation_string":"IBM Research,USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047310311","display_name":"Daniel Bola\u00f1os","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Daniel Bolanos","raw_affiliation_strings":["IBM Research,USA"],"affiliations":[{"raw_affiliation_string":"IBM Research,USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043856063","display_name":"Hyun Chae Jung","orcid":"https://orcid.org/0000-0001-5442-4341"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hyun Jung","raw_affiliation_strings":["IBM Research,USA"],"affiliations":[{"raw_affiliation_string":"IBM Research,USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5079994647","display_name":"George Saon","orcid":"https://orcid.org/0009-0004-6837-5009"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"George Saon","raw_affiliation_strings":["IBM Research,USA"],"affiliations":[{"raw_affiliation_string":"IBM Research,USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5069028016"],"corresponding_institution_ids":["https://openalex.org/I4210145865"],"apc_list":null,"apc_paid":null,"fwci":14.4724,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.98518131,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9506999850273132,"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":0.9506999850273132,"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/computer-science","display_name":"Computer science","score":0.7848626971244812},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6667901277542114},{"id":"https://openalex.org/keywords/resource","display_name":"Resource (disambiguation)","score":0.46446743607521057},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.40887847542762756},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.39572787284851074},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.08080396056175232}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7848626971244812},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6667901277542114},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.46446743607521057},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.40887847542762756},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39572787284851074},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.08080396056175232}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10888566","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888566","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":28,"referenced_works":["https://openalex.org/W1533416326","https://openalex.org/W2097003869","https://openalex.org/W2127141656","https://openalex.org/W2963211739","https://openalex.org/W2963240019","https://openalex.org/W2963490782","https://openalex.org/W2963626623","https://openalex.org/W2972574864","https://openalex.org/W3008037978","https://openalex.org/W3097777922","https://openalex.org/W3197140813","https://openalex.org/W3206573929","https://openalex.org/W4210811812","https://openalex.org/W4297841367","https://openalex.org/W4372259777","https://openalex.org/W4372260576","https://openalex.org/W4385807484","https://openalex.org/W4389519448","https://openalex.org/W4389520395","https://openalex.org/W4391021698","https://openalex.org/W4401042461","https://openalex.org/W4404783862","https://openalex.org/W6755207826","https://openalex.org/W6771467084","https://openalex.org/W6852909395","https://openalex.org/W6856530387","https://openalex.org/W6859665146","https://openalex.org/W6861626678"],"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/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Limited":[0],"transcribed":[1],"spoken":[2,70],"style":[3],"data":[4,33,65,78,122],"is":[5,35,90],"a":[6,20,118,141,185],"critical":[7],"bottleneck":[8],"in":[9,75,107,147],"building":[10,97],"automatic":[11],"speech":[12,143,165],"recognition":[13,144],"(ASR)":[14],"systems":[15,58],"for":[16,96,138,199],"low-resource":[17,56,109,119],"languages.":[18],"Prompting":[19],"large":[21],"language":[22,126],"model":[23,89,127,173],"(LLM)":[24],"to":[25,37,41,51,130,169,177],"paraphrase":[26],"input":[27],"text":[28,32,63,163],"can":[29],"generate":[30],"novel":[31],"that":[34],"constrained":[36],"be":[38],"semantically":[39],"similar":[40],"the":[42,53,61,68,76,84,87,108,124,132,157,161,171,180,193],"source":[43],"data.":[44],"We":[45,190],"leverage":[46],"this":[47,93],"capability":[48],"of":[49,55,86,104,149,188,195],"LLMs":[50],"improve":[52,179],"performance":[54],"ASR":[57,88,100],"by":[59,154,182],"increasing":[60],"limited":[62],"training":[64,77],"while":[66],"keeping":[67],"same":[69],"style.":[71],"Since":[72],"word":[73,150],"sequences":[74],"are":[79],"now":[80],"more":[81],"diverse":[82],"and":[83,166],"vocabulary":[85],"also":[91,191],"expanded,":[92],"approach":[94,198],"allows":[95],"general":[98],"purpose":[99],"without":[101],"prior":[102],"knowledge":[103],"various":[105],"domains":[106],"language.":[110],"In":[111],"our":[112,196],"experiments":[113],"with":[114,140],"Brazilian":[115],"Portuguese":[116],"as":[117],"language,":[120],"paraphrased":[121,162],"enhanced":[123],"n-gram":[125],"(LM)":[128],"used":[129],"build":[131],"weighted":[133],"finite":[134],"state":[135],"transducer":[136],"(WFST)":[137],"decoding":[139],"Conformer-CTC":[142],"model,":[145],"resulting":[146],"improvement":[148,187],"error":[151],"rate":[152],"(WER)":[153],"15.6%":[155],"over":[156],"baseline":[158],"model.":[159],"Synthesizing":[160],"into":[164],"using":[167],"it":[168],"fine-tune":[170],"acoustic":[172],"(AM)":[174],"component":[175],"helped":[176],"further":[178],"WER":[181],"2.9%,":[183],"achieving":[184],"combined":[186],"18.5%.":[189],"demonstrate":[192],"usefulness":[194],"proposed":[197],"high-resource":[200],"languages":[201],"like":[202],"English.":[203]},"counts_by_year":[{"year":2025,"cited_by_count":6}],"updated_date":"2026-04-03T22:45:19.894376","created_date":"2025-10-10T00:00:00"}