{"id":"https://openalex.org/W3142529478","doi":"https://doi.org/10.21437/interspeech.2021-555","title":"Multi-Encoder Learning and Stream Fusion for Transformer-Based End-to-End Automatic Speech Recognition","display_name":"Multi-Encoder Learning and Stream Fusion for Transformer-Based End-to-End Automatic Speech Recognition","publication_year":2021,"publication_date":"2021-08-27","ids":{"openalex":"https://openalex.org/W3142529478","doi":"https://doi.org/10.21437/interspeech.2021-555","mag":"3142529478"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2021-555","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-555","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2021","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2104.00120","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5039030019","display_name":"Timo Lohrenz","orcid":"https://orcid.org/0000-0001-7103-4302"},"institutions":[{"id":"https://openalex.org/I94509681","display_name":"Technische Universit\u00e4t Braunschweig","ror":"https://ror.org/010nsgg66","country_code":"DE","type":"education","lineage":["https://openalex.org/I94509681"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Timo Lohrenz","raw_affiliation_strings":["[Braunschweig University of Technology]"],"affiliations":[{"raw_affiliation_string":"[Braunschweig University of Technology]","institution_ids":["https://openalex.org/I94509681"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100736672","display_name":"Zhengyang Li","orcid":"https://orcid.org/0009-0007-6764-933X"},"institutions":[{"id":"https://openalex.org/I94509681","display_name":"Technische Universit\u00e4t Braunschweig","ror":"https://ror.org/010nsgg66","country_code":"DE","type":"education","lineage":["https://openalex.org/I94509681"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Zhengyang Li","raw_affiliation_strings":["[Braunschweig University of Technology]"],"affiliations":[{"raw_affiliation_string":"[Braunschweig University of Technology]","institution_ids":["https://openalex.org/I94509681"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002593702","display_name":"Tim Fingscheidt","orcid":"https://orcid.org/0000-0002-8895-5041"},"institutions":[{"id":"https://openalex.org/I94509681","display_name":"Technische Universit\u00e4t Braunschweig","ror":"https://ror.org/010nsgg66","country_code":"DE","type":"education","lineage":["https://openalex.org/I94509681"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Tim Fingscheidt","raw_affiliation_strings":["[Braunschweig University of Technology]"],"affiliations":[{"raw_affiliation_string":"[Braunschweig University of Technology]","institution_ids":["https://openalex.org/I94509681"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5039030019"],"corresponding_institution_ids":["https://openalex.org/I94509681"],"apc_list":null,"apc_paid":null,"fwci":0.699,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":{"value":0.75294731,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"2846","last_page":"2850"},"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":0.9998999834060669,"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":0.9991999864578247,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.728312611579895},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7034668922424316},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6319727897644043},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5652896165847778},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5643821954727173},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5233620405197144},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.51567542552948},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.4999735355377197},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.4789876937866211},{"id":"https://openalex.org/keywords/fusion","display_name":"Fusion","score":0.47736263275146484},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4701136648654938},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4213067293167114},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.38880884647369385},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.131756991147995}],"concepts":[{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.728312611579895},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7034668922424316},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6319727897644043},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5652896165847778},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5643821954727173},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5233620405197144},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.51567542552948},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.4999735355377197},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.4789876937866211},{"id":"https://openalex.org/C158525013","wikidata":"https://www.wikidata.org/wiki/Q2593739","display_name":"Fusion","level":2,"score":0.47736263275146484},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4701136648654938},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4213067293167114},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.38880884647369385},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.131756991147995},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","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/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},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.21437/interspeech.2021-555","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-555","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2021","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2104.00120","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2104.00120","pdf_url":"https://arxiv.org/pdf/2104.00120","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":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"mag:3142529478","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2104.00120.pdf","pdf_url":null,"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":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2104.00120","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2104.00120","pdf_url":null,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2104.00120","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2104.00120","pdf_url":"https://arxiv.org/pdf/2104.00120","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":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.5699999928474426,"id":"https://metadata.un.org/sdg/16"}],"awards":[{"id":"https://openalex.org/G3353889084","display_name":null,"funder_award_id":"414091002","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G5106512922","display_name":null,"funder_award_id":"Deutsche Forschungsgemeinschaft (DFG","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"}],"funders":[{"id":"https://openalex.org/F4320320879","display_name":"Deutsche Forschungsgemeinschaft","ror":"https://ror.org/018mejw64"},{"id":"https://openalex.org/F4320323803","display_name":"Bundesministerium f\u00fcr Wirtschaft und Energie","ror":"https://ror.org/02vgg2808"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3142529478.pdf","grobid_xml":"https://content.openalex.org/works/W3142529478.grobid-xml"},"referenced_works_count":47,"referenced_works":["https://openalex.org/W158407117","https://openalex.org/W202767486","https://openalex.org/W1494198834","https://openalex.org/W1524333225","https://openalex.org/W1553004968","https://openalex.org/W1576996184","https://openalex.org/W1904457459","https://openalex.org/W1915251500","https://openalex.org/W1929897159","https://openalex.org/W1993721840","https://openalex.org/W2024490156","https://openalex.org/W2095705004","https://openalex.org/W2096391593","https://openalex.org/W2102113734","https://openalex.org/W2103621378","https://openalex.org/W2112796928","https://openalex.org/W2143612262","https://openalex.org/W2168709618","https://openalex.org/W2303024254","https://openalex.org/W2327501763","https://openalex.org/W2526425061","https://openalex.org/W2626778328","https://openalex.org/W2787248535","https://openalex.org/W2793348910","https://openalex.org/W2885185669","https://openalex.org/W2886180730","https://openalex.org/W2890952074","https://openalex.org/W2892009249","https://openalex.org/W2900091092","https://openalex.org/W2913057142","https://openalex.org/W2933138175","https://openalex.org/W2936774411","https://openalex.org/W2948210185","https://openalex.org/W2952998108","https://openalex.org/W2962712961","https://openalex.org/W2962826786","https://openalex.org/W2963785710","https://openalex.org/W2964227577","https://openalex.org/W2970971581","https://openalex.org/W2972389417","https://openalex.org/W2972818416","https://openalex.org/W2974321942","https://openalex.org/W3007433671","https://openalex.org/W3008912312","https://openalex.org/W3035629723","https://openalex.org/W3094989221","https://openalex.org/W3096318498"],"related_works":["https://openalex.org/W3005525822","https://openalex.org/W2899640612","https://openalex.org/W3016010944","https://openalex.org/W2981696585","https://openalex.org/W2963590118","https://openalex.org/W2951378253","https://openalex.org/W2564155414","https://openalex.org/W2894651928","https://openalex.org/W3007433671","https://openalex.org/W3109079702","https://openalex.org/W3095913803","https://openalex.org/W2946555236","https://openalex.org/W3094929291","https://openalex.org/W3037029661","https://openalex.org/W2900092534","https://openalex.org/W3168770049","https://openalex.org/W3036175506","https://openalex.org/W2982413405","https://openalex.org/W3045904949","https://openalex.org/W3148654612"],"abstract_inverted_index":{"Stream":[0],"fusion,":[1],"also":[2],"known":[3,44],"as":[4,45],"system":[5],"combination,":[6],"is":[7],"a":[8,73,80,132,148],"common":[9],"technique":[10],"in":[11,58,99,121,136],"automatic":[12],"speech":[13],"recognition":[14],"for":[15,25,39,142],"traditional":[16],"hybrid":[17],"hidden":[18],"Markov":[19],"model":[20,31,115],"approaches,":[21],"yet":[22],"mostly":[23],"unexplored":[24],"modern":[26],"deep":[27],"neural":[28],"network":[29],"end-to-end":[30],"architectures.":[32],"Here,":[33],"we":[34,101,138],"investigate":[35],"various":[36],"fusion":[37,52,56,64,135],"techniques":[38],"the":[40,46,95,157],"all-attention-based":[41],"encoder-decoder":[42,85],"architecture":[43],"transformer,":[47],"striving":[48],"to":[49,104,156],"achieve":[50,139],"optimal":[51],"by":[53,131],"investigating":[54],"different":[55],"levels":[57],"an":[59],"example":[60],"single-microphone":[61],"setting":[62],"with":[63,113,147],"of":[65,83,152],"standard":[66],"magnitude":[67,96],"and":[68,116],"phase":[69],"features.":[70],"We":[71],"introduce":[72],"novel":[74],"multi-encoder":[75,128],"learning":[76],"method":[77],"that":[78],"performs":[79],"weighted":[81],"combination":[82],"two":[84,126],"multi-head":[86],"attention":[87],"outputs":[88],"only":[89,94],"during":[90],"training.":[91],"Employing":[92],"then":[93],"feature":[97],"encoder":[98],"inference,":[100,137],"are":[102],"able":[103],"show":[105],"consistent":[106],"improvement":[107],"on":[108,117,145],"Wall":[109],"Street":[110],"Journal":[111],"(WSJ)":[112],"language":[114],"Librispeech,":[118],"without":[119],"increase":[120],"runtime":[122],"or":[123],"parameters.":[124],"Combining":[125],"such":[127],"trained":[129],"models":[130,144],"simple":[133],"late":[134],"state-of-the-art":[140],"performance":[141],"transformer-based":[143],"WSJ":[146],"significant":[149],"WER":[150],"reduction":[151],"19%":[153],"relative":[154],"compared":[155],"current":[158],"benchmark":[159],"approach.":[160]},"counts_by_year":[{"year":2021,"cited_by_count":5}],"updated_date":"2026-03-12T06:13:28.667946","created_date":"2025-10-10T00:00:00"}