{"id":"https://openalex.org/W4212772209","doi":"https://doi.org/10.1109/lsp.2022.3152128","title":"Hybrid Autoregressive and Non-Autoregressive Transformer Models for Speech Recognition","display_name":"Hybrid Autoregressive and Non-Autoregressive Transformer Models for Speech Recognition","publication_year":2022,"publication_date":"2022-01-01","ids":{"openalex":"https://openalex.org/W4212772209","doi":"https://doi.org/10.1109/lsp.2022.3152128"},"language":"en","primary_location":{"id":"doi:10.1109/lsp.2022.3152128","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2022.3152128","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Signal Processing Letters","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2104.01522","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101817056","display_name":"Zhengkun Tian","orcid":"https://orcid.org/0000-0002-0469-3049"},"institutions":[{"id":"https://openalex.org/I4210100255","display_name":"Beijing Academy of Artificial Intelligence","ror":"https://ror.org/016a74861","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210100255"]},{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhengkun Tian","raw_affiliation_strings":["NLPR Lab, CASIA, Beijing, China","School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-0469-3049","affiliations":[{"raw_affiliation_string":"NLPR Lab, CASIA, Beijing, China","institution_ids":["https://openalex.org/I4210112150"]},{"raw_affiliation_string":"School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210100255","https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078525423","display_name":"Jiangyan Yi","orcid":"https://orcid.org/0000-0003-2422-4618"},"institutions":[{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiangyan Yi","raw_affiliation_strings":["NLPR Lab, CASIA, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-2422-4618","affiliations":[{"raw_affiliation_string":"NLPR Lab, CASIA, Beijing, China","institution_ids":["https://openalex.org/I4210112150"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112613657","display_name":"Jianhua Tao","orcid":"https://orcid.org/0000-0002-9344-6428"},"institutions":[{"id":"https://openalex.org/I4210097554","display_name":"Center for Excellence in Brain Science and Intelligence Technology","ror":"https://ror.org/00vpwhm04","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210097554"]},{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jianhua Tao","raw_affiliation_strings":["NLPR Lab, CASIA, Beijing, China","CAS Center for Excellence in Brain Science and Intelligence Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-9344-6428","affiliations":[{"raw_affiliation_string":"NLPR Lab, CASIA, Beijing, China","institution_ids":["https://openalex.org/I4210112150"]},{"raw_affiliation_string":"CAS Center for Excellence in Brain Science and Intelligence Technology, Beijing, China","institution_ids":["https://openalex.org/I4210097554"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Shuai Zhang","orcid":"https://orcid.org/0000-0002-1094-887X"},"institutions":[{"id":"https://openalex.org/I4210100255","display_name":"Beijing Academy of Artificial Intelligence","ror":"https://ror.org/016a74861","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210100255"]},{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shuai Zhang","raw_affiliation_strings":["NLPR Lab, CASIA, Beijing, China","School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-1094-887X","affiliations":[{"raw_affiliation_string":"NLPR Lab, CASIA, Beijing, China","institution_ids":["https://openalex.org/I4210112150"]},{"raw_affiliation_string":"School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210100255","https://openalex.org/I4210165038"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5111856667","display_name":"Zhengqi Wen","orcid":null},"institutions":[{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhengqi Wen","raw_affiliation_strings":["NLPR Lab, CASIA, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NLPR Lab, CASIA, Beijing, China","institution_ids":["https://openalex.org/I4210112150"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.7509,"has_fulltext":false,"cited_by_count":24,"citation_normalized_percentile":{"value":0.91350399,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":"29","issue":null,"first_page":"762","last_page":"766"},"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":0.9991999864578247,"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":0.9991000294685364,"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/autoregressive-model","display_name":"Autoregressive model","score":0.9253115057945251},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7733163833618164},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6134964227676392},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.59070885181427},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5875474810600281},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4518943130970001},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.45076972246170044},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.35524696111679077},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3549196124076843},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1159837543964386},{"id":"https://openalex.org/keywords/econometrics","display_name":"Econometrics","score":0.09700170159339905},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08051064610481262}],"concepts":[{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.9253115057945251},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7733163833618164},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6134964227676392},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.59070885181427},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5875474810600281},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4518943130970001},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.45076972246170044},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35524696111679077},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3549196124076843},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1159837543964386},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.09700170159339905},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08051064610481262},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/lsp.2022.3152128","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2022.3152128","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Signal Processing Letters","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2104.01522","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2104.01522","pdf_url":"https://arxiv.org/pdf/2104.01522","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2104.01522","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2104.01522","pdf_url":"https://arxiv.org/pdf/2104.01522","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5381013183","display_name":null,"funder_award_id":"61831022","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8374701036","display_name":null,"funder_award_id":"61901473","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W854541894","https://openalex.org/W1828163288","https://openalex.org/W2144499799","https://openalex.org/W2327501763","https://openalex.org/W2526425061","https://openalex.org/W2567070169","https://openalex.org/W2892009249","https://openalex.org/W2936078256","https://openalex.org/W2936774411","https://openalex.org/W2946200149","https://openalex.org/W2962760690","https://openalex.org/W2973122799","https://openalex.org/W2976556660","https://openalex.org/W2989134874","https://openalex.org/W3011339933","https://openalex.org/W3016010032","https://openalex.org/W3093941495","https://openalex.org/W3096160024","https://openalex.org/W3097625183","https://openalex.org/W3097874139","https://openalex.org/W3097882114","https://openalex.org/W3112157188","https://openalex.org/W3125815078","https://openalex.org/W3160622492","https://openalex.org/W3162431424","https://openalex.org/W3162899666","https://openalex.org/W4285789737","https://openalex.org/W4287628715","https://openalex.org/W4288088457","https://openalex.org/W4385245566","https://openalex.org/W6731370813","https://openalex.org/W6739901393","https://openalex.org/W6763832098","https://openalex.org/W6769806307","https://openalex.org/W6790121257"],"related_works":["https://openalex.org/W2053269318","https://openalex.org/W2546021431","https://openalex.org/W2581127593","https://openalex.org/W2136763963","https://openalex.org/W2109705048","https://openalex.org/W2940588515","https://openalex.org/W1909151225","https://openalex.org/W1987783679","https://openalex.org/W2160030256","https://openalex.org/W1521297879"],"abstract_inverted_index":{"The":[0,37,170],"autoregressive":[1,106],"(AR)":[2],"models,":[3],"such":[4],"as":[5],"attention-based":[6],"encoder-decoder":[7],"models":[8,40,83,97,117],"and":[9,28,52,84,100,107,115,138,164,185],"RNN-Transducer,":[10],"have":[11],"achieved":[12],"great":[13,76],"success":[14],"in":[15,58,78],"speech":[16],"recognition.":[17],"They":[18],"predict":[19,53],"the":[20,25,45,49,54,63,81,85,95,125,130,140,143,151,155,174,182,192,198],"output":[21,50,56],"sequence":[22],"conditioned":[23],"on":[24,35,159],"previous":[26],"tokens":[27,51,57],"acoustic":[29],"encoded":[30],"states,":[31],"which":[32,112],"is":[33,73,147,194],"inefficient":[34],"GPUs.":[36],"non-autoregressive":[38,108],"(NAR)":[39],"can":[41,176],"get":[42],"rid":[43],"of":[44,94,197],"temporal":[46],"dependency":[47],"between":[48,80],"entire":[55],"one":[59],"inference":[60,146],"step.":[61],"However,":[62],"NAR":[64,82,96,116,131,189],"model":[65,127,132,152,184],"still":[66,74],"faces":[67],"two":[68],"major":[69],"problems.":[70],"Firstly,":[71],"there":[72],"a":[75,104,160,165,178],"gap":[77],"performance":[79,180],"advanced":[86],"AR":[87,114,126,183,199],"models.":[88,190],"Secondly,":[89],"it\u2019s":[90],"difficult":[91],"for":[92],"most":[93],"to":[98,133,149],"train":[99],"converge.":[101],"We":[102,122],"propose":[103],"hybrid":[105,145],"transformer":[109],"(HANAT)":[110],"model,":[111],"integrates":[113],"deeply":[118],"by":[119],"sharing":[120],"parameters.":[121],"assume":[123],"that":[124,173],"will":[128],"assist":[129],"learn":[134],"some":[135],"linguistic":[136],"dependencies":[137],"accelerate":[139],"convergence.":[141],"Furthermore,":[142],"two-stage":[144],"applied":[148],"improve":[150],"performance.":[153],"All":[154],"experiments":[156],"are":[157],"conducted":[158],"mandarin":[161],"dataset":[162,167],"ASIEHLL-1":[163],"english":[166],"librispeech-960":[168],"h.":[169],"results":[171],"show":[172],"HANAT":[175],"achieve":[177],"competitive":[179],"with":[181],"outperform":[186],"many":[187],"complicated":[188],"Besides,":[191],"RTF":[193],"only":[195],"1/5":[196],"model.":[200]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":10},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":1}],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2022-02-24T00:00:00"}