{"id":"https://openalex.org/W3016010944","doi":"https://doi.org/10.1109/icassp40776.2020.9053455","title":"A Practical Two-Stage Training Strategy for Multi-Stream End-to-End Speech Recognition","display_name":"A Practical Two-Stage Training Strategy for Multi-Stream End-to-End Speech Recognition","publication_year":2020,"publication_date":"2020-04-09","ids":{"openalex":"https://openalex.org/W3016010944","doi":"https://doi.org/10.1109/icassp40776.2020.9053455","mag":"3016010944"},"language":"en","primary_location":{"id":"doi:10.1109/icassp40776.2020.9053455","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp40776.2020.9053455","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2020 - 2020 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/A5100760516","display_name":"Ruizhi Li","orcid":"https://orcid.org/0000-0002-2496-5224"},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ruizhi Li","raw_affiliation_strings":["Center for Language and Speech Processing, The Johns Hopkins University, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Language and Speech Processing, The Johns Hopkins University, USA","institution_ids":["https://openalex.org/I145311948"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046596198","display_name":"Gregory Sell","orcid":null},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gregory Sell","raw_affiliation_strings":["Center for Language and Speech Processing, The Johns Hopkins University, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Language and Speech Processing, The Johns Hopkins University, USA","institution_ids":["https://openalex.org/I145311948"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100363489","display_name":"Xiaofei Wang","orcid":"https://orcid.org/0009-0004-6683-3969"},"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":"Xiaofei Wang","raw_affiliation_strings":["Speech and Dialog Research Group, Microsoft, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Speech and Dialog Research Group, Microsoft, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001291873","display_name":"Shinji Watanabe","orcid":"https://orcid.org/0000-0002-5970-8631"},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shinji Watanabe","raw_affiliation_strings":["Center for Language and Speech Processing, The Johns Hopkins University, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Language and Speech Processing, The Johns Hopkins University, USA","institution_ids":["https://openalex.org/I145311948"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5042260050","display_name":"Hynek He\u0159mansk\u00fd","orcid":"https://orcid.org/0000-0001-8032-4811"},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hynek Hermansky","raw_affiliation_strings":["Center for Language and Speech Processing, The Johns Hopkins University, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Language and Speech Processing, The Johns Hopkins University, USA","institution_ids":["https://openalex.org/I145311948"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.92,"has_fulltext":false,"cited_by_count":8,"citation_normalized_percentile":{"value":0.78123765,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"7014","last_page":"7018"},"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":0.9997000098228455,"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"}},{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9994000196456909,"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":0.8281701803207397},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.7428117990493774},{"id":"https://openalex.org/keywords/data-stream","display_name":"Data stream","score":0.6106625199317932},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.5983919501304626},{"id":"https://openalex.org/keywords/extractor","display_name":"Extractor","score":0.5648393630981445},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.5424725413322449},{"id":"https://openalex.org/keywords/joint","display_name":"Joint (building)","score":0.5077760815620422},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4850313067436218},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.47151386737823486},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.45865437388420105},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.44878458976745605},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.44867607951164246},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.43723464012145996},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4150680899620056},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.06804677844047546}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8281701803207397},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.7428117990493774},{"id":"https://openalex.org/C2778484313","wikidata":"https://www.wikidata.org/wiki/Q1172540","display_name":"Data stream","level":2,"score":0.6106625199317932},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.5983919501304626},{"id":"https://openalex.org/C117978034","wikidata":"https://www.wikidata.org/wiki/Q5422192","display_name":"Extractor","level":2,"score":0.5648393630981445},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.5424725413322449},{"id":"https://openalex.org/C18555067","wikidata":"https://www.wikidata.org/wiki/Q8375051","display_name":"Joint (building)","level":2,"score":0.5077760815620422},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4850313067436218},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.47151386737823486},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.45865437388420105},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44878458976745605},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.44867607951164246},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.43723464012145996},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4150680899620056},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.06804677844047546},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"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/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0},{"id":"https://openalex.org/C21880701","wikidata":"https://www.wikidata.org/wiki/Q2144042","display_name":"Process engineering","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},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"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/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C170154142","wikidata":"https://www.wikidata.org/wiki/Q150737","display_name":"Architectural engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp40776.2020.9053455","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp40776.2020.9053455","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":43,"referenced_works":["https://openalex.org/W854541894","https://openalex.org/W1828163288","https://openalex.org/W1904457459","https://openalex.org/W2102113734","https://openalex.org/W2125336414","https://openalex.org/W2127141656","https://openalex.org/W2143612262","https://openalex.org/W2144121180","https://openalex.org/W2148613904","https://openalex.org/W2327501763","https://openalex.org/W2397147568","https://openalex.org/W2526425061","https://openalex.org/W2581377246","https://openalex.org/W2587464121","https://openalex.org/W2627092829","https://openalex.org/W2766219058","https://openalex.org/W2886180730","https://openalex.org/W2889508099","https://openalex.org/W2894835365","https://openalex.org/W2899640612","https://openalex.org/W2900070823","https://openalex.org/W2900091092","https://openalex.org/W2900092534","https://openalex.org/W2904113148","https://openalex.org/W2936774411","https://openalex.org/W2962780374","https://openalex.org/W2962866381","https://openalex.org/W2963064586","https://openalex.org/W2963211739","https://openalex.org/W2964012862","https://openalex.org/W2964391865","https://openalex.org/W4297798436","https://openalex.org/W6623517193","https://openalex.org/W6638749077","https://openalex.org/W6675365184","https://openalex.org/W6678809451","https://openalex.org/W6702097088","https://openalex.org/W6728030952","https://openalex.org/W6732696085","https://openalex.org/W6755235688","https://openalex.org/W6755820020","https://openalex.org/W6756137172","https://openalex.org/W6757106768"],"related_works":["https://openalex.org/W2151749779","https://openalex.org/W3179968364","https://openalex.org/W3082848404","https://openalex.org/W1979583797","https://openalex.org/W2016864125","https://openalex.org/W1999612375","https://openalex.org/W2372254676","https://openalex.org/W2938107654","https://openalex.org/W4390516098","https://openalex.org/W2080135837"],"abstract_inverted_index":{"The":[0],"multi-stream":[1,121,152],"paradigm":[2],"of":[3,61,68,80,167],"audio":[4],"processing,":[5],"in":[6,64],"which":[7],"several":[8,172],"sources":[9],"are":[10,108],"simultaneously":[11],"considered,":[12],"has":[13],"been":[14,142],"an":[15,58,65],"active":[16],"research":[17],"area":[18],"for":[19,83],"information":[20,41],"fusion.":[21],"Our":[22],"previous":[23,71,157],"study":[24],"offered":[25],"a":[26,44,91,100,111,120,151],"promising":[27],"direction":[28],"within":[29],"end-to-end":[30],"automatic":[31],"speech":[32],"recognition,":[33],"where":[34,105],"parallel":[35,81],"encoders":[36],"aim":[37],"to":[38,51,98,124],"capture":[39],"diverse":[40],"followed":[42],"by":[43],"stream-level":[45],"fusion":[46,129],"based":[47],"on":[48,144],"attention":[49,128],"mechanisms":[50],"combine":[52],"the":[53,70,127,132],"different":[54],"views.":[55],"However,":[56],"with":[57,115,155],"increasing":[59,66],"number":[60,67],"streams":[62],"resulting":[63],"encoders,":[69],"approach":[72],"could":[73],"require":[74],"substantial":[75],"memory":[76],"and":[77,135,148],"massive":[78],"amounts":[79],"data":[82],"joint":[84],"training.":[85],"In":[86],"this":[87,159],"work,":[88],"we":[89],"propose":[90],"practical":[92],"two-stage":[93],"training":[94],"scheme.":[95],"Stage-1":[96],"is":[97],"train":[99,126],"Universal":[101],"Feature":[102],"Extractor":[103],"(UFE),":[104],"encoder":[106],"outputs":[107],"produced":[109],"from":[110,138],"single-stream":[112],"model":[113],"trained":[114],"all":[116],"data.":[117],"Stage-2":[118],"formulates":[119],"scheme":[122],"intending":[123],"solely":[125],"module":[130],"using":[131],"UFE":[133],"features":[134],"pretrained":[136],"components":[137],"Stage-1.":[139],"Experiments":[140],"have":[141],"conducted":[143],"two":[145],"datasets,":[146],"DIRHA":[147],"AMI,":[149],"as":[150],"scenario.":[153],"Compared":[154],"our":[156],"method,":[158],"strategy":[160],"achieves":[161],"relative":[162],"word":[163],"error":[164],"rate":[165],"reductions":[166],"8.2-32.4%,":[168],"while":[169],"consistently":[170],"outperforming":[171],"conventional":[173],"combination":[174],"methods.":[175]},"counts_by_year":[{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":4},{"year":2020,"cited_by_count":3}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}