{"id":"https://openalex.org/W3095311338","doi":"https://doi.org/10.21437/interspeech.2020-2846","title":"On the Comparison of Popular End-to-End Models for Large Scale Speech Recognition","display_name":"On the Comparison of Popular End-to-End Models for Large Scale Speech Recognition","publication_year":2020,"publication_date":"2020-10-25","ids":{"openalex":"https://openalex.org/W3095311338","doi":"https://doi.org/10.21437/interspeech.2020-2846","mag":"3095311338"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2020-2846","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2020-2846","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2020","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/A5100365053","display_name":"Jinyu Li","orcid":"https://orcid.org/0000-0002-1089-9748"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Jinyu Li","raw_affiliation_strings":["Microsoft Speech and Language Group"],"affiliations":[{"raw_affiliation_string":"Microsoft Speech and Language Group","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101709477","display_name":"Yu Wu","orcid":"https://orcid.org/0000-0002-5715-3011"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu Wu","raw_affiliation_strings":["Microsoft Research Asia"],"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034136587","display_name":"Yashesh Gaur","orcid":null},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yashesh Gaur","raw_affiliation_strings":["Microsoft Speech and Language Group"],"affiliations":[{"raw_affiliation_string":"Microsoft Speech and Language Group","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101863385","display_name":"Chengyi Wang","orcid":"https://orcid.org/0000-0002-6780-9299"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chengyi Wang","raw_affiliation_strings":["Microsoft Research Asia"],"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101414593","display_name":"Rui Zhao","orcid":"https://orcid.org/0000-0002-5147-2695"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rui Zhao","raw_affiliation_strings":["Microsoft Speech and Language Group"],"affiliations":[{"raw_affiliation_string":"Microsoft Speech and Language Group","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101635405","display_name":"Shujie Liu","orcid":"https://orcid.org/0009-0008-0785-8882"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shujie Liu","raw_affiliation_strings":["Microsoft Research Asia"],"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5100365053"],"corresponding_institution_ids":["https://openalex.org/I1290206253"],"apc_list":null,"apc_paid":null,"fwci":14.1355,"has_fulltext":false,"cited_by_count":127,"citation_normalized_percentile":{"value":0.99136451,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":100},"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":0.9990000128746033,"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":0.9990000128746033,"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":0.9595000147819519,"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":0.9532999992370605,"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/end-to-end-principle","display_name":"End-to-end principle","score":0.823724627494812},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6890163421630859},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6018623113632202},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.5036525130271912},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3461947441101074}],"concepts":[{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.823724627494812},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6890163421630859},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6018623113632202},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.5036525130271912},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3461947441101074},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2020-2846","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2020-2846","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2020","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W854541894","https://openalex.org/W1686810756","https://openalex.org/W1736701665","https://openalex.org/W1828163288","https://openalex.org/W2102113734","https://openalex.org/W2133564696","https://openalex.org/W2160815625","https://openalex.org/W2194775991","https://openalex.org/W2327501763","https://openalex.org/W2396384435","https://openalex.org/W2605141709","https://openalex.org/W2750499125","https://openalex.org/W2892009249","https://openalex.org/W2902290872","https://openalex.org/W2939248538","https://openalex.org/W2962760690","https://openalex.org/W2962826786","https://openalex.org/W2963414781","https://openalex.org/W2963827914","https://openalex.org/W2972818416","https://openalex.org/W2973122799","https://openalex.org/W2987019345","https://openalex.org/W3007227084","https://openalex.org/W3008898571","https://openalex.org/W3015315932","https://openalex.org/W3015654466","https://openalex.org/W3016010032","https://openalex.org/W3016155933","https://openalex.org/W3094667432","https://openalex.org/W3095189764","https://openalex.org/W3097777922","https://openalex.org/W4294619417","https://openalex.org/W4302296459","https://openalex.org/W4385245566","https://openalex.org/W4394666973"],"related_works":["https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W3179968364","https://openalex.org/W2390279801","https://openalex.org/W3196421258","https://openalex.org/W2358668433","https://openalex.org/W2376932109","https://openalex.org/W2151749779","https://openalex.org/W2382290278","https://openalex.org/W2938107654"],"abstract_inverted_index":{"Recently,":[0],"there":[1,20],"has":[2,97],"been":[3,98],"a":[4,156],"strong":[5],"push":[6],"to":[7,12,65,78],"transition":[8],"from":[9],"hybrid":[10,158],"models":[11,15,70,105,150],"end-to-end":[13],"(E2E)":[14],"for":[16],"automatic":[17],"speech":[18],"recognition.Currently,":[19],"are":[21,71,106],"three":[22,129],"promising":[23],"E2E":[24,69,130],"methods:":[25],"recurrent":[26],"neural":[27],"network":[28],"transducer":[29],"(RNN-T),":[30],"RNN":[31],"attentionbased":[32],"encoder-decoder":[33],"(AED),":[34],"and":[35,47,53,140,148],"Transformer-AED.In":[36],"this":[37],"study,":[38],"we":[39],"conduct":[40],"an":[41],"empirical":[42],"comparison":[43],"of":[44,60,85,90],"RNN-T,":[45],"RNN-AED,":[46],"Transformer-AED":[48],"models,":[49,131],"in":[50,110,118,137],"both":[51,138,145],"non-streaming":[52,112,141],"streaming":[54,119,139,146],"modes.We":[55],"use":[56],"65":[57],"thousand":[58],"hours":[59],"Microsoft":[61],"anonymized":[62],"training":[63,86],"data":[64,73],"train":[66],"these":[67],"models.As":[68],"more":[72],"hungry,":[74],"it":[75],"is":[76,115],"better":[77,153],"compare":[79],"their":[80],"effectiveness":[81],"with":[82],"large":[83],"amount":[84],"data.To":[87],"the":[88,111,134],"best":[89,135],"our":[91],"knowledge,":[92],"no":[93],"such":[94],"comprehensive":[95],"study":[96],"conducted":[99],"yet.We":[100],"show":[101,143],"that":[102,144],"although":[103],"AED":[104],"stronger":[107],"than":[108,155],"RNN-T":[109,114,147],"mode,":[113],"very":[116],"competitive":[117],"mode":[120],"if":[121],"its":[122],"encoder":[123],"can":[124,151],"be":[125],"properly":[126],"initialized.Among":[127],"all":[128],"transformer-AED":[132,149],"achieved":[133],"accuracy":[136,154],"mode.We":[142],"obtain":[152],"highly-optimized":[157],"model.":[159]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":7},{"year":2024,"cited_by_count":14},{"year":2023,"cited_by_count":27},{"year":2022,"cited_by_count":23},{"year":2021,"cited_by_count":42},{"year":2020,"cited_by_count":12}],"updated_date":"2026-04-18T07:56:08.524223","created_date":"2025-10-10T00:00:00"}