{"id":"https://openalex.org/W3016072087","doi":"https://doi.org/10.1109/icassp40776.2020.9054245","title":"Structured Sparse Attention for end-to-end Automatic Speech Recognition","display_name":"Structured Sparse Attention for end-to-end Automatic Speech Recognition","publication_year":2020,"publication_date":"2020-04-09","ids":{"openalex":"https://openalex.org/W3016072087","doi":"https://doi.org/10.1109/icassp40776.2020.9054245","mag":"3016072087"},"language":"en","primary_location":{"id":"doi:10.1109/icassp40776.2020.9054245","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp40776.2020.9054245","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":"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/A5015690293","display_name":"Jiabin Xue","orcid":null},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Jiabin Xue","raw_affiliation_strings":["School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China"],"affiliations":[{"raw_affiliation_string":"School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101846141","display_name":"Tieran Zheng","orcid":"https://orcid.org/0009-0006-5179-0278"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tieran Zheng","raw_affiliation_strings":["School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China"],"affiliations":[{"raw_affiliation_string":"School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101508830","display_name":"Jiqing Han","orcid":"https://orcid.org/0000-0002-4297-4300"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiqing Han","raw_affiliation_strings":["School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China"],"affiliations":[{"raw_affiliation_string":"School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China","institution_ids":["https://openalex.org/I204983213"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5015690293"],"corresponding_institution_ids":["https://openalex.org/I204983213"],"apc_list":null,"apc_paid":null,"fwci":0.2651,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.6142882,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":"15","issue":null,"first_page":"7044","last_page":"7048"},"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":0.9993000030517578,"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":0.998199999332428,"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":0.8560543656349182},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6998679637908936},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6121298670768738},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.36925601959228516}],"concepts":[{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.8560543656349182},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6998679637908936},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6121298670768738},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36925601959228516}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp40776.2020.9054245","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp40776.2020.9054245","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":[{"score":0.5299999713897705,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W6908809","https://openalex.org/W370143576","https://openalex.org/W854541894","https://openalex.org/W1494198834","https://openalex.org/W1524333225","https://openalex.org/W2095705004","https://openalex.org/W2127141656","https://openalex.org/W2193413348","https://openalex.org/W2253795368","https://openalex.org/W2327501763","https://openalex.org/W2417863416","https://openalex.org/W2766219058","https://openalex.org/W2889163603","https://openalex.org/W2892009249","https://openalex.org/W2962780374","https://openalex.org/W2962824709","https://openalex.org/W2962826786","https://openalex.org/W2963123301","https://openalex.org/W2963403664","https://openalex.org/W2963403868","https://openalex.org/W2964330921","https://openalex.org/W4385245566","https://openalex.org/W4396738453","https://openalex.org/W6600284362","https://openalex.org/W6623517193","https://openalex.org/W6629717138","https://openalex.org/W6631362777","https://openalex.org/W6674330103","https://openalex.org/W6687566353","https://openalex.org/W6691766336","https://openalex.org/W6739901393","https://openalex.org/W6754299077"],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W3179968364","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W2376932109","https://openalex.org/W1999612375","https://openalex.org/W2001405890","https://openalex.org/W2382290278","https://openalex.org/W2938107654","https://openalex.org/W2478288626"],"abstract_inverted_index":{"The":[0,172],"Softmax":[1,99],"normalization":[2],"function-based":[3],"attention":[4,33,73,86,106,143],"mechanism":[5,29,87],"is":[6,51,169],"often":[7],"employed":[8],"by":[9,88,108,150],"End-to-End":[10],"Automatic":[11],"Speech":[12],"Recognition":[13],"(E2E":[14],"ASR)":[15],"models":[16],"to":[17,22,31,59,83,96,141,144,183],"tell":[18],"the":[19,25,32,39,45,61,85,90,98,101,105,110,114,119,122,139,152,167],"network":[20],"where":[21],"focus":[23],"within":[24],"input.":[26],"However,":[27],"this":[28,49,67],"leads":[30],"distribution":[34],"becoming":[35],"increasingly":[36],"flatter":[37],"as":[38],"input":[40],"sequence":[41],"length":[42],"increases,":[43],"since":[44],"output":[46],"probability":[47,115],"of":[48,148],"function":[50],"dense":[52],"and":[53],"nonnegative,":[54],"which":[55,81],"makes":[56],"it":[57],"unable":[58],"highlight":[60],"important":[62],"information":[63],"in":[64,113,166],"speech.":[65],"In":[66],"paper,":[68],"we":[69,94,130],"present":[70,132],"two":[71],"sparse":[72,91,135],"mechanisms":[74],"for":[75],"ASR":[76,180,192],"tasks":[77],"with":[78,100],"long":[79],"utterances,":[80],"try":[82],"improve":[84],"introducing":[89],"transformation.":[92],"First,":[93],"propose":[95],"replace":[97],"Sparsemax":[102],"that":[103,137,162,175],"normalizes":[104],"weight":[107],"finding":[109],"closest":[111],"point":[112],"simplex.":[116],"Then,":[117],"considering":[118],"structured":[120,134],"characteristics,":[121],"pronunciation":[123],"has":[124],"a":[125,133,145,184,190],"relatively":[126],"stable":[127],"duration.":[128],"Therefore,":[129],"further":[131],"transformation":[136],"forces":[138],"networks":[140],"pay":[142],"continuous":[146],"segment":[147],"speech":[149],"applying":[151],"l":[153],"<sub":[154],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[155],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">2</sub>":[156],"penalty.":[157],"A":[158],"noniterative":[159],"solution":[160],"algorithm":[161],"can":[163],"be":[164],"used":[165],"backpropagation":[168],"designed":[170],"here.":[171],"experiments":[173],"show":[174],"our":[176],"methods":[177],"achieve":[178],"better":[179],"results":[181],"compared":[182],"well-tuned":[185],"attention-based":[186],"baseline":[187],"system":[188],"on":[189],"character":[191],"task.":[193]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}