{"id":"https://openalex.org/W4225808286","doi":"https://doi.org/10.1109/asru51503.2021.9688232","title":"Adapting GPT, GPT-2 and BERT Language Models for Speech Recognition","display_name":"Adapting GPT, GPT-2 and BERT Language Models for Speech Recognition","publication_year":2021,"publication_date":"2021-12-13","ids":{"openalex":"https://openalex.org/W4225808286","doi":"https://doi.org/10.1109/asru51503.2021.9688232"},"language":"en","primary_location":{"id":"doi:10.1109/asru51503.2021.9688232","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru51503.2021.9688232","pdf_url":null,"source":{"id":"https://openalex.org/S4363606113","display_name":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","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/A5013736436","display_name":"Xianrui Zheng","orcid":null},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Xianrui Zheng","raw_affiliation_strings":["Cambridge University,Engineering Dept.,Cambridge,U.K.,CB2 1PZ"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cambridge University,Engineering Dept.,Cambridge,U.K.,CB2 1PZ","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100460206","display_name":"Chao Zhang","orcid":"https://orcid.org/0000-0002-7730-5131"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Chao Zhang","raw_affiliation_strings":["Cambridge University,Engineering Dept.,Cambridge,U.K.,CB2 1PZ"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cambridge University,Engineering Dept.,Cambridge,U.K.,CB2 1PZ","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002191410","display_name":"Philip C. Woodland","orcid":"https://orcid.org/0000-0001-9069-0225"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Philip C. Woodland","raw_affiliation_strings":["Cambridge University,Engineering Dept.,Cambridge,U.K.,CB2 1PZ"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Cambridge University,Engineering Dept.,Cambridge,U.K.,CB2 1PZ","institution_ids":["https://openalex.org/I241749"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I241749"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":76,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"162","last_page":"168"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998000264167786,"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":0.9998000264167786,"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/T10028","display_name":"Topic Modeling","score":0.9994000196456909,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9994000196456909,"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":0.7751119136810303},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.6703444719314575},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6037547588348389},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5867249369621277},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5212962627410889},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.47615906596183777},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.45366790890693665},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.36049431562423706},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.0741424560546875}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7751119136810303},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.6703444719314575},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6037547588348389},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5867249369621277},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5212962627410889},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.47615906596183777},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45366790890693665},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.36049431562423706},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0741424560546875},{"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/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asru51503.2021.9688232","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru51503.2021.9688232","pdf_url":null,"source":{"id":"https://openalex.org/S4363606113","display_name":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.8199999928474426,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":72,"referenced_works":["https://openalex.org/W101286142","https://openalex.org/W179875071","https://openalex.org/W1508165687","https://openalex.org/W1522301498","https://openalex.org/W1524333225","https://openalex.org/W1566289585","https://openalex.org/W1970689298","https://openalex.org/W2058695628","https://openalex.org/W2076094076","https://openalex.org/W2112036188","https://openalex.org/W2133564696","https://openalex.org/W2155388323","https://openalex.org/W2160451571","https://openalex.org/W2171928131","https://openalex.org/W2275115695","https://openalex.org/W2402268235","https://openalex.org/W2513257850","https://openalex.org/W2514741789","https://openalex.org/W2747236259","https://openalex.org/W2747917286","https://openalex.org/W2888867175","https://openalex.org/W2889177077","https://openalex.org/W2896457183","https://openalex.org/W2943845043","https://openalex.org/W2962739339","https://openalex.org/W2962784628","https://openalex.org/W2963026768","https://openalex.org/W2963654251","https://openalex.org/W2964110616","https://openalex.org/W2965373594","https://openalex.org/W2970597249","https://openalex.org/W2979826702","https://openalex.org/W2990391581","https://openalex.org/W2996428491","https://openalex.org/W2998704965","https://openalex.org/W3015484572","https://openalex.org/W3024308166","https://openalex.org/W3034775979","https://openalex.org/W3094841848","https://openalex.org/W3096471021","https://openalex.org/W3141961557","https://openalex.org/W3180465246","https://openalex.org/W4288089799","https://openalex.org/W4292779060","https://openalex.org/W4295312788","https://openalex.org/W4299838440","https://openalex.org/W4303633609","https://openalex.org/W4385245566","https://openalex.org/W6631190155","https://openalex.org/W6631362777","https://openalex.org/W6665295360","https://openalex.org/W6669713995","https://openalex.org/W6679434410","https://openalex.org/W6680532216","https://openalex.org/W6684821475","https://openalex.org/W6694643851","https://openalex.org/W6725816475","https://openalex.org/W6739901393","https://openalex.org/W6742632731","https://openalex.org/W6743280976","https://openalex.org/W6750253784","https://openalex.org/W6754111305","https://openalex.org/W6755207826","https://openalex.org/W6763701032","https://openalex.org/W6766673545","https://openalex.org/W6766978945","https://openalex.org/W6768021236","https://openalex.org/W6769627184","https://openalex.org/W6770528390","https://openalex.org/W6778883912","https://openalex.org/W6780226713","https://openalex.org/W6798187346"],"related_works":["https://openalex.org/W2309273277","https://openalex.org/W4288365749","https://openalex.org/W2936497627","https://openalex.org/W3013624417","https://openalex.org/W2061937230","https://openalex.org/W4287826556","https://openalex.org/W1574295218","https://openalex.org/W3098382480","https://openalex.org/W4287598411","https://openalex.org/W3100913109"],"abstract_inverted_index":{"Language":[0],"models":[1],"(LMs)":[2],"pre-trained":[3],"on":[4,88,99,131,147],"massive":[5],"amounts":[6],"of":[7,63,112,121,170],"text,":[8],"in":[9,92,176],"particular":[10],"bidirectional":[11,59,89],"encoder":[12],"representations":[13],"from":[14,129],"Transformers":[15],"(BERT),":[16],"generative":[17],"pre-training":[18],"(GPT),":[19],"and":[20,43,55,104,116,167,173],"GPT-2,":[21,42,56],"have":[22],"become":[23],"a":[24,70,93,138],"key":[25],"technology":[26],"for":[27,46,154],"many":[28],"natural":[29],"language":[30,72,84,155],"processing":[31],"tasks.":[32],"In":[33],"this":[34],"paper,":[35],"we":[36],"present":[37],"results":[38,98,175],"using":[39],"fine-tuned":[40,114],"GPT,":[41],"their":[44],"combination":[45,111,120,169],"automatic":[47],"speech":[48],"recognition":[49],"(ASR).":[50],"Unlike":[51],"unidirectional":[52],"LM":[53,90],"GPT":[54,115,172],"BERT":[57,159],"is":[58,67,78],"whose":[60],"direct":[61],"product":[62],"the":[64,82,100,110,113,119,132,148,151,168],"output":[65],"probabilities":[66,157],"no":[68],"longer":[69],"valid":[71],"prior":[73,85,156],"probability.":[74],"A":[75],"conversion":[76,153],"method":[77],"proposed":[79,152],"to":[80,137,160],"compute":[81],"correct":[83],"probability":[86],"based":[87],"outputs":[91],"mathematically":[94],"exact":[95],"way.":[96],"Experimental":[97],"widely":[101],"used":[102],"AMI":[103,149],"Switchboard":[105],"ASR":[106],"tasks":[107],"showed":[108],"that":[109],"GPT-2":[117,174],"outperformed":[118],"three":[122],"neural":[123],"LMs":[124],"with":[125],"different":[126],"architectures":[127],"trained":[128],"scratch":[130],"in-domain":[133],"text":[134],"by":[135],"up":[136],"12%":[139],"relative":[140,165],"word":[141],"error":[142],"rate":[143],"reduction":[144],"(WERR).":[145],"Furthermore,":[146],"corpus,":[150],"enables":[158],"obtain":[161],"an":[162],"extra":[163],"3%":[164],"WERR,":[166],"BERT,":[171],"further":[177],"improvements.":[178]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":28},{"year":2024,"cited_by_count":23},{"year":2023,"cited_by_count":21},{"year":2022,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}