{"id":"https://openalex.org/W3093712593","doi":"https://doi.org/10.1109/icassp39728.2021.9414353","title":"Transformer-Based End-to-End Speech Recognition with Local Dense Synthesizer Attention","display_name":"Transformer-Based End-to-End Speech Recognition with Local Dense Synthesizer Attention","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3093712593","doi":"https://doi.org/10.1109/icassp39728.2021.9414353","mag":"3093712593"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9414353","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414353","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2010.12155","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102318256","display_name":"Menglong Xu","orcid":"https://orcid.org/0009-0006-0878-5502"},"institutions":[{"id":"https://openalex.org/I17145004","display_name":"Northwestern Polytechnical University","ror":"https://ror.org/01y0j0j86","country_code":"CN","type":"education","lineage":["https://openalex.org/I17145004"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Menglong Xu","raw_affiliation_strings":["School of Marine Science and Technology, Northwestern Polytechnical University, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Marine Science and Technology, Northwestern Polytechnical University, China","institution_ids":["https://openalex.org/I17145004"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101523840","display_name":"Shengqiang Li","orcid":"https://orcid.org/0000-0002-0837-6366"},"institutions":[{"id":"https://openalex.org/I17145004","display_name":"Northwestern Polytechnical University","ror":"https://ror.org/01y0j0j86","country_code":"CN","type":"education","lineage":["https://openalex.org/I17145004"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shengqiang Li","raw_affiliation_strings":["School of Marine Science and Technology, Northwestern Polytechnical University, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Marine Science and Technology, Northwestern Polytechnical University, China","institution_ids":["https://openalex.org/I17145004"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100450091","display_name":"Xiao-Lei Zhang","orcid":"https://orcid.org/0000-0001-7694-193X"},"institutions":[{"id":"https://openalex.org/I17145004","display_name":"Northwestern Polytechnical University","ror":"https://ror.org/01y0j0j86","country_code":"CN","type":"education","lineage":["https://openalex.org/I17145004"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiao-Lei Zhang","raw_affiliation_strings":["School of Marine Science and Technology, Northwestern Polytechnical University, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Marine Science and Technology, Northwestern Polytechnical University, China","institution_ids":["https://openalex.org/I17145004"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":4.058,"has_fulltext":false,"cited_by_count":36,"citation_normalized_percentile":{"value":0.94673643,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"5899","last_page":"5903"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":1.0,"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":1.0,"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":0.9994999766349792,"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":0.9979000091552734,"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":0.7627182602882385},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.7277534604072571},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6157352328300476},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5171908140182495},{"id":"https://openalex.org/keywords/pairwise-comparison","display_name":"Pairwise comparison","score":0.47045958042144775},{"id":"https://openalex.org/keywords/computational-complexity-theory","display_name":"Computational complexity theory","score":0.44307276606559753},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3946264684200287},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.1858464479446411},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.1189805269241333},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.10820251703262329},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.08470186591148376}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7627182602882385},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7277534604072571},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6157352328300476},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5171908140182495},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.47045958042144775},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.44307276606559753},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3946264684200287},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.1858464479446411},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.1189805269241333},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.10820251703262329},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.08470186591148376}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp39728.2021.9414353","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414353","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2010.12155","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2010.12155","pdf_url":"https://arxiv.org/pdf/2010.12155","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:2010.12155","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2010.12155","pdf_url":"https://arxiv.org/pdf/2010.12155","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":[{"score":0.4099999964237213,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320329139","display_name":"Communication University of China","ror":"https://ror.org/04facbs33"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W2193413348","https://openalex.org/W2327501763","https://openalex.org/W2514741789","https://openalex.org/W2766219058","https://openalex.org/W2892009249","https://openalex.org/W2908336025","https://openalex.org/W2936774411","https://openalex.org/W2939111082","https://openalex.org/W2952809536","https://openalex.org/W2962778134","https://openalex.org/W2963240019","https://openalex.org/W2963242190","https://openalex.org/W2963403868","https://openalex.org/W2963827914","https://openalex.org/W2964272710","https://openalex.org/W2972818416","https://openalex.org/W2976556660","https://openalex.org/W2981857663","https://openalex.org/W2982413405","https://openalex.org/W3006683367","https://openalex.org/W3006801027","https://openalex.org/W3007328579","https://openalex.org/W3015966793","https://openalex.org/W3015974384","https://openalex.org/W3016010032","https://openalex.org/W3021293129","https://openalex.org/W3024732798","https://openalex.org/W3025165719","https://openalex.org/W3097777922","https://openalex.org/W3097874139","https://openalex.org/W3106504817","https://openalex.org/W4288088457","https://openalex.org/W4294619417","https://openalex.org/W4385245566","https://openalex.org/W6687566353","https://openalex.org/W6739901393","https://openalex.org/W6746574493","https://openalex.org/W6757585730","https://openalex.org/W6769806307","https://openalex.org/W6774054309","https://openalex.org/W6776684981"],"related_works":["https://openalex.org/W2487162673","https://openalex.org/W2793211469","https://openalex.org/W2949152769","https://openalex.org/W4372354731","https://openalex.org/W2807634898","https://openalex.org/W1692008701","https://openalex.org/W2942366970","https://openalex.org/W2597588799","https://openalex.org/W4360593462","https://openalex.org/W2562400057"],"abstract_inverted_index":{"Recently,":[0],"several":[1],"studies":[2],"reported":[3],"that":[4,21,115,132],"dot-product":[5],"self-attention":[6],"(SA)":[7],"may":[8],"not":[9,149],"be":[10],"indispensable":[11],"to":[12,55,72,79,97],"the":[13,19,59,64,74,84,99,108,116,134,137,143,159,165,174,179],"state-of-the-art":[14],"Transformer":[15],"models.":[16],"Motivated":[17],"by":[18],"fact":[20],"dense":[22],"synthesizer":[23],"attention":[24,75],"(DSA),":[25],"which":[26,127,156],"dispenses":[27],"with":[28,95],"dot":[29],"products":[30],"and":[31,62,101,170],"pairwise":[32],"interactions,":[33],"achieved":[34],"competitive":[35],"results":[36,106],"in":[37,42],"many":[38],"language":[39],"processing":[40],"tasks,":[41],"this":[43],"paper,":[44],"we":[45,66,92],"first":[46],"propose":[47,68],"a":[48,80,120,152],"DSA-based":[49],"speech":[50,89,111],"recognition,":[51],"as":[52,173],"an":[53],"alternative":[54],"SA.":[56],"To":[57],"reduce":[58],"computational":[60,171],"complexity":[61,172],"improve":[63],"performance,":[65],"further":[67],"local":[69,81,100],"DSA":[70,78],"(LDSA)":[71],"restrict":[73],"scope":[76],"of":[77,125,133,154,168,178],"range":[82],"around":[83],"current":[85],"central":[86],"frame":[87],"for":[88],"recognition.":[90],"Finally,":[91],"combine":[93],"LDSA":[94,181],"SA":[96],"extract":[98],"global":[102],"information":[103],"simultaneously.":[104],"Experimental":[105],"on":[107],"Ai-shell1":[109],"Mandarin":[110],"recognition":[112],"corpus":[113],"show":[114],"proposed":[117,146],"LDSA-Transformer":[118,138],"achieves":[119,151],"character":[121],"error":[122],"rate":[123],"(CER)":[124],"6.49%,":[126],"is":[128,182],"slightly":[129],"better":[130],"than":[131,142],"SA-Transformer.":[135,144],"Meanwhile,":[136],"requires":[139],"less":[140],"computation":[141],"The":[145,176],"combination":[147],"method":[148],"only":[150],"CER":[153],"6.18%,":[155],"significantly":[157],"outperforms":[158],"SA-Transformer,":[160],"but":[161],"also":[162],"has":[163],"roughly":[164],"same":[166],"number":[167],"parameters":[169],"latter.":[175],"implementation":[177],"multi-head":[180],"available":[183],"at":[184],"https://github.com/mlxu995/multihead-LDSA":[185]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":9},{"year":2022,"cited_by_count":8},{"year":2021,"cited_by_count":7},{"year":2020,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2020-10-29T00:00:00"}