{"id":"https://openalex.org/W4372259951","doi":"https://doi.org/10.1109/icassp49357.2023.10095330","title":"Self-Convolution for Automatic Speech Recognition","display_name":"Self-Convolution for Automatic Speech Recognition","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372259951","doi":"https://doi.org/10.1109/icassp49357.2023.10095330"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10095330","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10095330","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 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/A5100669565","display_name":"Tianhao Zhang","orcid":"https://orcid.org/0000-0002-5939-3932"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Tian-Hao Zhang","raw_affiliation_strings":["University of Science and Technology Beijing,Department of Computer Science and Technology,Beijing,China,100083","USTB-EEasyTech Joint Lab of Artificial Intelligence, University of Science and Technology Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,Department of Computer Science and Technology,Beijing,China,100083","institution_ids":["https://openalex.org/I92403157"]},{"raw_affiliation_string":"USTB-EEasyTech Joint Lab of Artificial Intelligence, University of Science and Technology Beijing, Beijing, China","institution_ids":["https://openalex.org/I92403157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100453144","display_name":"Qi Liu","orcid":"https://orcid.org/0000-0001-5378-6404"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qi Liu","raw_affiliation_strings":["University of Science and Technology Beijing,Department of Computer Science and Technology,Beijing,China,100083","USTB-EEasyTech Joint Lab of Artificial Intelligence, University of Science and Technology Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,Department of Computer Science and Technology,Beijing,China,100083","institution_ids":["https://openalex.org/I92403157"]},{"raw_affiliation_string":"USTB-EEasyTech Joint Lab of Artificial Intelligence, University of Science and Technology Beijing, Beijing, China","institution_ids":["https://openalex.org/I92403157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056495776","display_name":"Xinyuan Qian","orcid":"https://orcid.org/0000-0002-9511-6713"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinyuan Qian","raw_affiliation_strings":["University of Science and Technology Beijing,Department of Computer Science and Technology,Beijing,China,100083"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,Department of Computer Science and Technology,Beijing,China,100083","institution_ids":["https://openalex.org/I92403157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006074341","display_name":"Song-Lu Chen","orcid":"https://orcid.org/0000-0002-0780-658X"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Song-Lu Chen","raw_affiliation_strings":["University of Science and Technology Beijing,Department of Computer Science and Technology,Beijing,China,100083","USTB-EEasyTech Joint Lab of Artificial Intelligence, University of Science and Technology Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,Department of Computer Science and Technology,Beijing,China,100083","institution_ids":["https://openalex.org/I92403157"]},{"raw_affiliation_string":"USTB-EEasyTech Joint Lab of Artificial Intelligence, University of Science and Technology Beijing, Beijing, China","institution_ids":["https://openalex.org/I92403157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100352754","display_name":"Feng Chen","orcid":"https://orcid.org/0000-0003-4813-2494"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Feng Chen","raw_affiliation_strings":["University of Science and Technology Beijing,USTB-EEasyTech Joint Lab of Artificial Intelligence,Beijing,China,100083","EEasy Technology Company Ltd, Zhuhai, China"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,USTB-EEasyTech Joint Lab of Artificial Intelligence,Beijing,China,100083","institution_ids":["https://openalex.org/I92403157"]},{"raw_affiliation_string":"EEasy Technology Company Ltd, Zhuhai, China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5074514262","display_name":"Xu-Cheng Yin","orcid":"https://orcid.org/0000-0003-0023-0220"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu-Cheng Yin","raw_affiliation_strings":["University of Science and Technology Beijing,Department of Computer Science and Technology,Beijing,China,100083","USTB-EEasyTech Joint Lab of Artificial Intelligence, University of Science and Technology Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,Department of Computer Science and Technology,Beijing,China,100083","institution_ids":["https://openalex.org/I92403157"]},{"raw_affiliation_string":"USTB-EEasyTech Joint Lab of Artificial Intelligence, University of Science and Technology Beijing, Beijing, China","institution_ids":["https://openalex.org/I92403157"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5100669565"],"corresponding_institution_ids":["https://openalex.org/I92403157"],"apc_list":null,"apc_paid":null,"fwci":0.3491,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.63012806,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"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/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"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.998199999332428,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.79588782787323},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7818257212638855},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.7121657133102417},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.6117342710494995},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6000986099243164},{"id":"https://openalex.org/keywords/range","display_name":"Range (aeronautics)","score":0.5104502439498901},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4854632019996643},{"id":"https://openalex.org/keywords/contrast","display_name":"Contrast (vision)","score":0.42252117395401},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.40200597047805786},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.3439900875091553},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.27975356578826904},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11158174276351929}],"concepts":[{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.79588782787323},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7818257212638855},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.7121657133102417},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.6117342710494995},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6000986099243164},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.5104502439498901},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4854632019996643},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.42252117395401},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.40200597047805786},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3439900875091553},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.27975356578826904},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11158174276351929},{"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/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49357.2023.10095330","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10095330","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320336125","display_name":"National Science Fund for Distinguished Young Scholars","ror":null},{"id":"https://openalex.org/F4320337504","display_name":"Research and Development","ror":"https://ror.org/027s68j25"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":50,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1522301498","https://openalex.org/W1524333225","https://openalex.org/W2066561607","https://openalex.org/W2251321385","https://openalex.org/W2407080277","https://openalex.org/W2567070169","https://openalex.org/W2892009249","https://openalex.org/W2936774411","https://openalex.org/W2962778134","https://openalex.org/W2962784628","https://openalex.org/W2963242190","https://openalex.org/W2964110616","https://openalex.org/W3033529678","https://openalex.org/W3093712593","https://openalex.org/W3095173472","https://openalex.org/W3097580812","https://openalex.org/W3097777922","https://openalex.org/W3138516171","https://openalex.org/W3141854550","https://openalex.org/W3157506437","https://openalex.org/W3162090017","https://openalex.org/W3162249256","https://openalex.org/W3177349073","https://openalex.org/W3197522112","https://openalex.org/W3197813307","https://openalex.org/W3200006799","https://openalex.org/W4225525539","https://openalex.org/W4225829036","https://openalex.org/W4226504707","https://openalex.org/W4297775537","https://openalex.org/W4312443924","https://openalex.org/W4385245566","https://openalex.org/W6629717138","https://openalex.org/W6631190155","https://openalex.org/W6631362777","https://openalex.org/W6691770337","https://openalex.org/W6731370813","https://openalex.org/W6737664043","https://openalex.org/W6739901393","https://openalex.org/W6754299077","https://openalex.org/W6779163297","https://openalex.org/W6784012970","https://openalex.org/W6791038633","https://openalex.org/W6792376965","https://openalex.org/W6795140394","https://openalex.org/W6796417832","https://openalex.org/W6801596000","https://openalex.org/W6810370753","https://openalex.org/W6811121879"],"related_works":["https://openalex.org/W4388335561","https://openalex.org/W2970530566","https://openalex.org/W2964954556","https://openalex.org/W3034421924","https://openalex.org/W2982536526","https://openalex.org/W4386858688","https://openalex.org/W4380302312","https://openalex.org/W3008689640","https://openalex.org/W4385338604","https://openalex.org/W3081626085"],"abstract_inverted_index":{"Self-attention":[0],"plays":[1],"a":[2,65],"significant":[3],"role":[4],"in":[5,25,41,48,100],"recent":[6],"automatic":[7],"speech":[8],"recognition":[9],"(ASR)":[10],"models":[11],"with":[12,151],"promising":[13],"results.":[14],"However,":[15],"it":[16,46],"suffers":[17],"from":[18],"high":[19],"computational":[20,153],"complexity":[21],"and":[22,39,50,63,123,135,138],"weak":[23],"capability":[24],"modeling":[26],"local":[27,43,86],"information.":[28,44],"In":[29,116],"contrast,":[30],"the":[31,104,111,119],"convolutional":[32],"neural":[33],"network":[34],"(CNN)":[35],"is":[36],"computationally":[37],"effective":[38],"superior":[40,146],"learning":[42,112],"Whereas":[45],"fails":[47],"self-interaction":[49],"capturing":[51],"long-range":[52,114],"dependence":[53],"among":[54],"input":[55],"tokens.":[56],"Accordingly,":[57],"we":[58,98],"take":[59],"their":[60],"complementary":[61],"advantages":[62,120],"propose":[64],"new":[66],"module,":[67],"namely":[68],"self-convolution,":[69],"to":[70,92,109],"compensate":[71],"for":[72],"each":[73,82],"individual":[74],"limitations.":[75],"Specifically,":[76],"self-convolution":[77,143],"generates":[78],"convolution":[79,107],"kernels":[80],"at":[81],"token":[83],"(to":[84],"model":[85],"information)":[87],"which":[88],"are":[89,125],"then":[90],"used":[91],"convolve":[93],"itself":[94],"(for":[95],"self-interaction).":[96],"Moreover,":[97],"bring":[99],"global":[101],"information":[102],"during":[103],"generation":[105],"of":[106,113,121],"kernel":[108],"enhance":[110],"dependencies.":[115],"this":[117],"way,":[118],"self-attention":[122,150],"CNN":[124],"both":[126],"utilized.":[127],"We":[128],"conduct":[129],"rigorous":[130],"experiments":[131],"on":[132],"LibriSpeech,":[133],"Tedlium2,":[134],"AIShell1":[136],"datasets":[137],"demonstrate":[139],"that":[140],"our":[141],"proposed":[142],"can":[144],"achieve":[145],"ASR":[147],"performance":[148],"than":[149],"less":[152],"cost.":[154]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}