{"id":"https://openalex.org/W4392931212","doi":"https://doi.org/10.1109/icassp48485.2024.10448361","title":"Multilingual and Fully Non-Autoregressive ASR with Large Language Model Fusion: A Comprehensive Study","display_name":"Multilingual and Fully Non-Autoregressive ASR with Large Language Model Fusion: A Comprehensive Study","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392931212","doi":"https://doi.org/10.1109/icassp48485.2024.10448361"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10448361","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10448361","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5091738469","display_name":"W. Ronny Huang","orcid":"https://orcid.org/0009-0000-8136-5666"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"W. Ronny Huang","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030888546","display_name":"Cyril Allauzen","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Cyril Allauzen","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076026485","display_name":"Tongzhou Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tongzhou Chen","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028500687","display_name":"Kilol Gupta","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kilol Gupta","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100691022","display_name":"Ke Hu","orcid":"https://orcid.org/0000-0003-2078-9371"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ke Hu","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048771433","display_name":"James Qin","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"James Qin","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100433648","display_name":"Yu Zhang","orcid":"https://orcid.org/0000-0002-9505-1833"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yu Zhang","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100339989","display_name":"Yongqiang Wang","orcid":"https://orcid.org/0000-0002-8600-0276"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yongqiang Wang","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001306222","display_name":"Shuo-Yiin Chang","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shuo-Yiin Chang","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5070513394","display_name":"Tara N. Sainath","orcid":"https://orcid.org/0000-0002-4126-6556"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tara N. Sainath","raw_affiliation_strings":["Google Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"13306","last_page":"13310"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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":0.9998999834060669,"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":0.9995999932289124,"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/T11309","display_name":"Music and Audio Processing","score":0.9980000257492065,"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":0.8088866472244263},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.729152262210846},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.645073413848877},{"id":"https://openalex.org/keywords/vocabulary","display_name":"Vocabulary","score":0.6378740668296814},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.6324431300163269},{"id":"https://openalex.org/keywords/closed-captioning","display_name":"Closed captioning","score":0.6180017590522766},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5534159541130066},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.5238088369369507},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5009863376617432},{"id":"https://openalex.org/keywords/autoregressive-integrated-moving-average","display_name":"Autoregressive integrated moving average","score":0.44518420100212097},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4303748309612274},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38063710927963257},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3569033741950989},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.26735660433769226},{"id":"https://openalex.org/keywords/time-series","display_name":"Time series","score":0.15255919098854065},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.12337365746498108},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.11905255913734436},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.09977048635482788},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.08426204323768616}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8088866472244263},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.729152262210846},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.645073413848877},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.6378740668296814},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.6324431300163269},{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.6180017590522766},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5534159541130066},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5238088369369507},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5009863376617432},{"id":"https://openalex.org/C24338571","wikidata":"https://www.wikidata.org/wiki/Q2566298","display_name":"Autoregressive integrated moving average","level":3,"score":0.44518420100212097},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4303748309612274},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38063710927963257},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3569033741950989},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.26735660433769226},{"id":"https://openalex.org/C151406439","wikidata":"https://www.wikidata.org/wiki/Q186588","display_name":"Time series","level":2,"score":0.15255919098854065},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.12337365746498108},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.11905255913734436},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.09977048635482788},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.08426204323768616},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"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/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10448361","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10448361","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.5400000214576721,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W2513345070","https://openalex.org/W2786018489","https://openalex.org/W2963240019","https://openalex.org/W2973727699","https://openalex.org/W3008898571","https://openalex.org/W3024308166","https://openalex.org/W3034775979","https://openalex.org/W3096032230","https://openalex.org/W3097777922","https://openalex.org/W3149629662","https://openalex.org/W3164692279","https://openalex.org/W3204696009","https://openalex.org/W4210463634","https://openalex.org/W4224308101","https://openalex.org/W4224457787","https://openalex.org/W4226521565","https://openalex.org/W4283026156","https://openalex.org/W4288089799","https://openalex.org/W4292779060","https://openalex.org/W4319862635","https://openalex.org/W4322718191","https://openalex.org/W4323066695","https://openalex.org/W4360836968","https://openalex.org/W4372259777","https://openalex.org/W4372259817","https://openalex.org/W4372269772","https://openalex.org/W4372349408","https://openalex.org/W4385823247","https://openalex.org/W4391021457","https://openalex.org/W6755207826","https://openalex.org/W6767997687","https://openalex.org/W6769627184","https://openalex.org/W6778883912","https://openalex.org/W6810081322","https://openalex.org/W6838461927","https://openalex.org/W6847363464","https://openalex.org/W6850218400","https://openalex.org/W6850625674","https://openalex.org/W6850936240","https://openalex.org/W6852800892"],"related_works":["https://openalex.org/W4210416330","https://openalex.org/W2775506363","https://openalex.org/W3088136942","https://openalex.org/W4290852288","https://openalex.org/W2949362007","https://openalex.org/W4283207562","https://openalex.org/W2963177403","https://openalex.org/W2330246314","https://openalex.org/W2949522393","https://openalex.org/W4388747839"],"abstract_inverted_index":{"In":[0],"the":[1,6,30,39,45,94,114,117],"era":[2],"of":[3,9,33,63,96,119],"large":[4],"models,":[5],"autoregressive":[7],"nature":[8],"decoding":[10],"often":[11],"results":[12],"in":[13,50],"latency":[14],"serving":[15],"as":[16,81],"a":[17,22],"significant":[18],"bottleneck.":[19],"We":[20],"propose":[21],"non-autoregressive":[23],"LM-fused":[24,122],"ASR":[25,106],"system":[26],"that":[27],"effectively":[28],"leverages":[29],"parallelization":[31],"capabilities":[32],"accelerator":[34],"hardware.":[35],"Our":[36],"approach":[37],"combines":[38],"Universal":[40],"Speech":[41],"Model":[42],"(USM)":[43],"and":[44,67],"PaLM":[46],"2":[47],"language":[48],"model":[49],"per-segment":[51],"scoring":[52],"mode,":[53],"achieving":[54],"an":[55],"average":[56],"relative":[57],"WER":[58],"improvement":[59],"across":[60],"all":[61],"languages":[62],"10.8%":[64],"on":[65,69,105],"FLEURS":[66],"3.6%":[68],"YouTube":[70],"captioning.":[71],"Furthermore,":[72],"our":[73],"comprehensive":[74],"ablation":[75],"study":[76,109],"analyzes":[77],"key":[78],"parameters":[79,104],"such":[80],"LLM":[82,97],"size,":[83,87],"context":[84],"length,":[85],"vocabulary":[86],"fusion":[88],"methodology.":[89],"For":[90],"instance,":[91],"we":[92],"explore":[93],"impact":[95],"size":[98],"ranging":[99],"from":[100],"128M":[101],"to":[102],"340B":[103],"performance.":[107],"This":[108],"provides":[110],"valuable":[111],"insights":[112],"into":[113],"factors":[115],"influencing":[116],"effectiveness":[118],"practical":[120],"large-scale":[121],"speech":[123],"recognition":[124],"systems.":[125]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":6}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}