{"id":"https://openalex.org/W3015492205","doi":"https://doi.org/10.1109/icassp40776.2020.9053568","title":"Dfsmn-San with Persistent Memory Model for Automatic Speech Recognition","display_name":"Dfsmn-San with Persistent Memory Model for Automatic Speech Recognition","publication_year":2020,"publication_date":"2020-04-09","ids":{"openalex":"https://openalex.org/W3015492205","doi":"https://doi.org/10.1109/icassp40776.2020.9053568","mag":"3015492205"},"language":"en","primary_location":{"id":"doi:10.1109/icassp40776.2020.9053568","doi":"https://doi.org/10.1109/icassp40776.2020.9053568","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp40776.2020.9053568","pdf_url":null,"source":{"id":"https://openalex.org/S4363607879","display_name":"IEEE International Conference on Acoustics Speech and Signal Processing","issn_l":"1520-6149","issn":["1520-6149","2379-190X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["jufo-2","norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"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"},"volume":null,"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/A5101967525","display_name":"Zhao You","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhao You","raw_affiliation_strings":["Tencent AI Lab, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent AI Lab, Shenzhen, China","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075183307","display_name":"Dan Su","orcid":"https://orcid.org/0000-0001-5746-9545","observed_orcids":["https://orcid.org/0000-0001-5746-9545"]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dan Su","raw_affiliation_strings":["Tencent AI Lab, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent AI Lab, Shenzhen, China","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100782730","display_name":"Jie Chen","orcid":"https://orcid.org/0000-0001-8419-4620","observed_orcids":["https://orcid.org/0000-0001-8419-4620","https://orcid.org/0000-0002-5994-2997"]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jie Chen","raw_affiliation_strings":["Tencent AI Lab, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent AI Lab, Shenzhen, China","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106404246","display_name":"Chao Weng","orcid":"https://orcid.org/0009-0009-8712-9176","observed_orcids":["https://orcid.org/0009-0009-8712-9176"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chao Weng","raw_affiliation_strings":["Tencent AI Lab, Bellevue, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent AI Lab, Bellevue, WA, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5034476404","display_name":"Dong Yu","orcid":"https://orcid.org/0000-0003-0520-6844","observed_orcids":["https://orcid.org/0000-0003-0520-6844"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dong Yu","raw_affiliation_strings":["Tencent AI Lab, Bellevue, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent AI Lab, Bellevue, WA, USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.5116,"has_fulltext":false,"cited_by_count":16,"citation_normalized_percentile":{"value":0.86122946,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"7704","last_page":"7708"},"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/T10860","display_name":"Speech and Audio Processing","score":9.999999747378752e-05,"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/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/self-attention-network","display_name":"self-attention network","score":0.8450000286102295},{"id":"https://openalex.org/keywords/persistent-memory","display_name":"persistent memory","score":0.8420000076293945},{"id":"https://openalex.org/keywords/long-term-dependencies","display_name":"long-range dependencies","score":0.8209999799728394},{"id":"https://openalex.org/keywords/large-vocabulary-continuous-speech-recognition","display_name":"large vocabulary continuous speech recognition","score":0.8539999723434448},{"id":"https://openalex.org/keywords/character-error-rate","display_name":"character error rate","score":0.6169999837875366},{"id":"https://openalex.org/keywords/encoder-architecture","display_name":"encoder architecture","score":0.5239999890327454}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8639034628868103},{"id":"https://openalex.org/C2775852435","wikidata":"https://www.wikidata.org/wiki/Q258403","display_name":"Utterance","level":2,"score":0.7747373580932617},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5901302099227905},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.5335559844970703},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.5215887427330017},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5169236063957214},{"id":"https://openalex.org/C133488467","wikidata":"https://www.wikidata.org/wiki/Q6673524","display_name":"Long short term memory","level":4,"score":0.499448299407959},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.47426116466522217},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4695587456226349},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4280984401702881},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.33399903774261475},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.3180011808872223},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.2668950855731964},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"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/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp40776.2020.9053568","doi":"https://doi.org/10.1109/icassp40776.2020.9053568","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp40776.2020.9053568","pdf_url":null,"source":{"id":"https://openalex.org/S4363607879","display_name":"IEEE International Conference on Acoustics Speech and Signal Processing","issn_l":"1520-6149","issn":["1520-6149","2379-190X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["jufo-2","norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"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":[],"sustainable_development_goals_aurora":[{"id":"https://metadata.un.org/sdg/4","score":0.4399999976158142,"display_name":"Quality education"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W1489048302","https://openalex.org/W1524333225","https://openalex.org/W2127141656","https://openalex.org/W2405883473","https://openalex.org/W2626778328","https://openalex.org/W2775304348","https://openalex.org/W2786031273","https://openalex.org/W2795138957","https://openalex.org/W2802023636","https://openalex.org/W2892009249","https://openalex.org/W2911291251","https://openalex.org/W2913718171","https://openalex.org/W2955227499","https://openalex.org/W2962778134","https://openalex.org/W2962824709","https://openalex.org/W2963308316","https://openalex.org/W2964089206","https://openalex.org/W4288289156","https://openalex.org/W4385245566","https://openalex.org/W6631362777","https://openalex.org/W6739901393","https://openalex.org/W6765264507"],"related_works":["https://openalex.org/W2912153778","https://openalex.org/W4288108708","https://openalex.org/W4387163678","https://openalex.org/W2973430807","https://openalex.org/W4385280324","https://openalex.org/W2890685186","https://openalex.org/W2984436043","https://openalex.org/W4390245176","https://openalex.org/W3173606726","https://openalex.org/W2904359912"],"abstract_inverted_index":{"Self-attention":[0],"networks":[1],"(SAN)":[2],"have":[3],"been":[4],"introduced":[5],"into":[6,119],"automatic":[7],"speech":[8],"recognition":[9],"(ASR)":[10],"and":[11,63,110],"achieved":[12],"state-of-the-art":[13],"performance":[14],"owing":[15],"to":[16,49,67,74,95,158],"its":[17],"superior":[18],"ability":[19],"in":[20,144,162],"capturing":[21],"long":[22],"term":[23],"dependency.":[24],"One":[25],"of":[26,114],"the":[27,31,40,56,134,151],"key":[28],"ingredients":[29],"is":[30],"self-attention":[32,69,93,105,120],"mechanism":[33],"which":[34,84],"can":[35,60],"be":[36,61],"effectively":[37],"performed":[38],"on":[39,123,129],"whole":[41,57],"utterance":[42,58],"level.":[43],"In":[44],"this":[45],"paper,":[46],"we":[47,77,108],"try":[48],"investigate":[50],"whether":[51],"even":[52],"more":[53],"information":[54],"beyond":[55],"level":[59],"exploited":[62],"beneficial.":[64],"We":[65],"propose":[66,79,109],"apply":[68],"layer":[70],"with":[71,92,102],"augmented":[72],"memory":[73,89,116,153],"ASR.":[75],"Specifically,":[76],"first":[78],"a":[80,97,103],"variant":[81],"model":[82,100],"architecture":[83,136],"combines":[85],"deep":[86],"feed-forward":[87],"sequential":[88],"network":[90],"(DFSMN)":[91],"layers":[94],"form":[96],"better":[98],"baseline":[99],"compared":[101],"purely":[104],"network.":[106],"Then,":[107],"compare":[111],"two":[112],"kinds":[113],"additional":[115,152],"structures":[117],"added":[118],"layers.":[121],"Experiments":[122],"large-scale":[124],"LVCSR":[125],"tasks":[126],"show":[127],"that":[128],"four":[130],"individual":[131],"test":[132],"sets,":[133],"DFSMN-SAN":[135],"outperforms":[137],"vanilla":[138],"SAN":[139],"encoder":[140],"by":[141],"5%":[142,157],"relatively":[143],"character":[145],"error":[146],"rate":[147],"(CER).":[148],"More":[149],"importantly,":[150],"structure":[154],"provides":[155],"further":[156],"11%":[159],"relative":[160],"improvement":[161],"CER.":[163]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":7},{"year":2020,"cited_by_count":5}],"updated_date":"2026-10-12T09:40:04.860316","created_date":"2025-10-10T00:00:00"}