{"id":"https://openalex.org/W3141464856","doi":"https://doi.org/10.1109/slt48900.2021.9383497","title":"Transformer Based Deliberation for Two-Pass Speech Recognition","display_name":"Transformer Based Deliberation for Two-Pass Speech Recognition","publication_year":2021,"publication_date":"2021-01-19","ids":{"openalex":"https://openalex.org/W3141464856","doi":"https://doi.org/10.1109/slt48900.2021.9383497","mag":"3141464856"},"language":"en","primary_location":{"id":"doi:10.1109/slt48900.2021.9383497","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt48900.2021.9383497","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE Spoken Language Technology Workshop (SLT)","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/A5029338576","display_name":"Ke Hu","orcid":"https://orcid.org/0000-0002-1599-1519"},"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":true,"raw_author_name":"Ke Hu","raw_affiliation_strings":["Google, Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google, Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112901893","display_name":"Ruoming Pang","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":"Ruoming Pang","raw_affiliation_strings":["Google, Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google, Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","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, Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google, Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5032931723","display_name":"Trevor Strohman","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":"Trevor Strohman","raw_affiliation_strings":["Google, Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google, Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5029338576"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":2.8553,"has_fulltext":false,"cited_by_count":24,"citation_normalized_percentile":{"value":0.92082627,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"68","last_page":"74"},"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/T11309","display_name":"Music and Audio Processing","score":0.9998000264167786,"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/T10860","display_name":"Speech and Audio Processing","score":0.9997000098228455,"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":0.8190665245056152},{"id":"https://openalex.org/keywords/deliberation","display_name":"Deliberation","score":0.811948299407959},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.7592366933822632},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.6246839761734009},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.609953761100769},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.5219842195510864},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.45564210414886475},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.36162516474723816},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.20789280533790588},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.08497121930122375}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8190665245056152},{"id":"https://openalex.org/C2776946740","wikidata":"https://www.wikidata.org/wiki/Q358652","display_name":"Deliberation","level":3,"score":0.811948299407959},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7592366933822632},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6246839761734009},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.609953761100769},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5219842195510864},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.45564210414886475},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36162516474723816},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.20789280533790588},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.08497121930122375},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"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/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/slt48900.2021.9383497","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt48900.2021.9383497","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16","score":0.5099999904632568}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":62,"referenced_works":["https://openalex.org/W97072897","https://openalex.org/W854541894","https://openalex.org/W1828163288","https://openalex.org/W2033256038","https://openalex.org/W2064675550","https://openalex.org/W2086161653","https://openalex.org/W2121879602","https://openalex.org/W2133564696","https://openalex.org/W2166637769","https://openalex.org/W2213952365","https://openalex.org/W2327501763","https://openalex.org/W2402144811","https://openalex.org/W2515439472","https://openalex.org/W2617258110","https://openalex.org/W2752047430","https://openalex.org/W2916997151","https://openalex.org/W2928941594","https://openalex.org/W2935756939","https://openalex.org/W2953384591","https://openalex.org/W2962760690","https://openalex.org/W2962824709","https://openalex.org/W2962893195","https://openalex.org/W2963149635","https://openalex.org/W2963403868","https://openalex.org/W2963747784","https://openalex.org/W2964308564","https://openalex.org/W2972818416","https://openalex.org/W2972880214","https://openalex.org/W2973122799","https://openalex.org/W2981857663","https://openalex.org/W2982413405","https://openalex.org/W3007328579","https://openalex.org/W3008181812","https://openalex.org/W3011339933","https://openalex.org/W3015194534","https://openalex.org/W3015506840","https://openalex.org/W3015723703","https://openalex.org/W3015726069","https://openalex.org/W3015752032","https://openalex.org/W3016010032","https://openalex.org/W3027687786","https://openalex.org/W3037029661","https://openalex.org/W3094667432","https://openalex.org/W3095311338","https://openalex.org/W3095994005","https://openalex.org/W3097973766","https://openalex.org/W4288088457","https://openalex.org/W4385245566","https://openalex.org/W6603931906","https://openalex.org/W6623517193","https://openalex.org/W6638749077","https://openalex.org/W6679434410","https://openalex.org/W6688428952","https://openalex.org/W6713134421","https://openalex.org/W6739901393","https://openalex.org/W6743726175","https://openalex.org/W6760633627","https://openalex.org/W6769557084","https://openalex.org/W6769806307","https://openalex.org/W6775002949","https://openalex.org/W6775605244","https://openalex.org/W6779924452"],"related_works":["https://openalex.org/W2753773138","https://openalex.org/W3204354834","https://openalex.org/W6292469","https://openalex.org/W2223796429","https://openalex.org/W2801092833","https://openalex.org/W2901193592","https://openalex.org/W156167186","https://openalex.org/W2050445137","https://openalex.org/W2221419418","https://openalex.org/W2981757109"],"abstract_inverted_index":{"Interactive":[0],"speech":[1],"recognition":[2],"systems":[3],"must":[4],"generate":[5],"words":[6],"quickly":[7,25],"while":[8],"also":[9,145],"producing":[10],"accurate":[11],"results.":[12],"Two-pass":[13],"models":[14],"excel":[15],"at":[16,62],"these":[17],"requirements":[18],"by":[19,118],"employing":[20],"a":[21,29,45,119,139,153],"first-pass":[22,73,108],"decoder":[23,31],"that":[24,32,44],"emits":[26],"words,":[27],"and":[28,67,107,151],"second-pass":[30,52],"requires":[33],"more":[34,38],"context":[35,113],"but":[36],"is":[37],"accurate.":[39],"Previous":[40],"work":[41],"has":[42],"established":[43],"deliberation":[46,91,129],"network":[47],"can":[48],"be":[49],"an":[50],"effective":[51],"model.":[53,74],"The":[54,111],"model":[55],"attends":[56],"to":[57,101,103,123],"two":[58],"kinds":[59],"of":[60,84],"inputs":[61],"once:":[63],"encoded":[64,105],"audio":[65,106],"frames":[66],"the":[68,72,98],"hypothesis":[69],"text":[70,109],"from":[71],"In":[75,93],"this":[76],"work,":[77],"we":[78,96],"explore":[79],"using":[80],"transformer":[81,94,128,149],"layers":[82,89],"instead":[83],"long-short":[85],"term":[86],"memory":[87],"(LSTM)":[88],"for":[90],"rescoring.":[92],"layers,":[95],"generalize":[97],"\"encoder-decoder\"":[99],"attention":[100],"attend":[102],"both":[104],"hypotheses.":[110],"output":[112],"vectors":[114],"are":[115],"then":[116],"combined":[117],"merger":[120],"layer.":[121],"Compared":[122],"LSTM-based":[124],"deliberation,":[125],"our":[126],"best":[127],"achieves":[130],"7%":[131],"relative":[132,155],"word":[133],"error":[134],"rate":[135],"improvements":[136],"along":[137],"with":[138],"38%":[140],"reduction":[141],"in":[142],"computation.":[143],"We":[144],"compare":[146],"against":[147],"non-deliberation":[148],"rescoring,":[150],"find":[152],"9%":[154],"improvement.":[156]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":11},{"year":2021,"cited_by_count":5}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}
