{"id":"https://openalex.org/W4319862646","doi":"https://doi.org/10.1109/slt54892.2023.10023451","title":"Macro-Block Dropout for Improved Regularization in Training End-to-End Speech Recognition Models","display_name":"Macro-Block Dropout for Improved Regularization in Training End-to-End Speech Recognition Models","publication_year":2023,"publication_date":"2023-01-09","ids":{"openalex":"https://openalex.org/W4319862646","doi":"https://doi.org/10.1109/slt54892.2023.10023451"},"language":"en","primary_location":{"id":"doi:10.1109/slt54892.2023.10023451","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/slt54892.2023.10023451","pdf_url":null,"source":{"id":"https://openalex.org/S4363605953","display_name":"2022 IEEE Spoken Language Technology Workshop (SLT)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"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":"2022 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"},"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/A5100684422","display_name":"Chanwoo Kim","orcid":"https://orcid.org/0000-0003-0193-8167"},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Chanwoo Kim","raw_affiliation_strings":["Samsung Research,Seoul,South Korea","Samsung Research, Seoul, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Research,Seoul,South Korea","institution_ids":["https://openalex.org/I2250650973"]},{"raw_affiliation_string":"Samsung Research, Seoul, South Korea","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032462992","display_name":"Sathish Indurti","orcid":null},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Sathish Indurti","raw_affiliation_strings":["Samsung Research,Seoul,South Korea","Samsung Research, Seoul, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Research,Seoul,South Korea","institution_ids":["https://openalex.org/I2250650973"]},{"raw_affiliation_string":"Samsung Research, Seoul, South Korea","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108758586","display_name":"Jin-Hwan Park","orcid":null},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jinhwan Park","raw_affiliation_strings":["Samsung Research,Seoul,South Korea","Samsung Research, Seoul, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Research,Seoul,South Korea","institution_ids":["https://openalex.org/I2250650973"]},{"raw_affiliation_string":"Samsung Research, Seoul, South Korea","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5113491293","display_name":"Wonyong Sung","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Wonyong Sung","raw_affiliation_strings":["Seoul National University,Seoul,South Korea","Seoul National University, Seoul, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University,Seoul,South Korea","institution_ids":["https://openalex.org/I139264467"]},{"raw_affiliation_string":"Seoul National University, Seoul, South Korea","institution_ids":["https://openalex.org/I139264467"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.00734598,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"28","issue":null,"first_page":"331","last_page":"338"},"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":0.9987999796867371,"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":0.9983000159263611,"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/dropout","display_name":"Dropout (neural networks)","score":0.7776041030883789},{"id":"https://openalex.org/keywords/overfitting","display_name":"Overfitting","score":0.6550272703170776},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6345012784004211},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.6115487217903137},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5092278718948364},{"id":"https://openalex.org/keywords/macro","display_name":"Macro","score":0.4519890248775482},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.44429516792297363},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.42977991700172424},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.39952516555786133},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3572845458984375},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.3442750573158264},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.285981684923172},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.24924463033676147}],"concepts":[{"id":"https://openalex.org/C2776145597","wikidata":"https://www.wikidata.org/wiki/Q25339462","display_name":"Dropout (neural networks)","level":2,"score":0.7776041030883789},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.6550272703170776},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6345012784004211},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.6115487217903137},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5092278718948364},{"id":"https://openalex.org/C166955791","wikidata":"https://www.wikidata.org/wiki/Q629579","display_name":"Macro","level":2,"score":0.4519890248775482},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.44429516792297363},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.42977991700172424},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39952516555786133},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3572845458984375},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3442750573158264},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.285981684923172},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.24924463033676147},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/slt54892.2023.10023451","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/slt54892.2023.10023451","pdf_url":null,"source":{"id":"https://openalex.org/S4363605953","display_name":"2022 IEEE Spoken Language Technology Workshop (SLT)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"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":"2022 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":50,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2062227835","https://openalex.org/W2063656732","https://openalex.org/W2064675550","https://openalex.org/W2095705004","https://openalex.org/W2096779346","https://openalex.org/W2105806712","https://openalex.org/W2107508685","https://openalex.org/W2127141656","https://openalex.org/W2139427956","https://openalex.org/W2143612262","https://openalex.org/W2150678643","https://openalex.org/W2292443655","https://openalex.org/W2326699523","https://openalex.org/W2396889049","https://openalex.org/W2559655401","https://openalex.org/W2592356860","https://openalex.org/W2604272474","https://openalex.org/W2612690371","https://openalex.org/W2617258110","https://openalex.org/W2747329762","https://openalex.org/W2785572076","https://openalex.org/W2899575547","https://openalex.org/W2902772991","https://openalex.org/W2903105043","https://openalex.org/W2936774411","https://openalex.org/W2962824709","https://openalex.org/W2963747784","https://openalex.org/W2963980003","https://openalex.org/W3007185811","https://openalex.org/W3008587939","https://openalex.org/W3015823121","https://openalex.org/W3025075133","https://openalex.org/W3162308011","https://openalex.org/W3168770049","https://openalex.org/W4212774754","https://openalex.org/W6600213771","https://openalex.org/W6623517193","https://openalex.org/W6631190155","https://openalex.org/W6638545294","https://openalex.org/W6674330103","https://openalex.org/W6684191040","https://openalex.org/W6688167117","https://openalex.org/W6713134421","https://openalex.org/W6714181750","https://openalex.org/W6739901393","https://openalex.org/W6754484989","https://openalex.org/W6755651582","https://openalex.org/W6756690837","https://openalex.org/W6768214904"],"related_works":["https://openalex.org/W3128220493","https://openalex.org/W4298017035","https://openalex.org/W2792147139","https://openalex.org/W4226354336","https://openalex.org/W3110700750","https://openalex.org/W2998675825","https://openalex.org/W2736804899","https://openalex.org/W2897443685","https://openalex.org/W4307654087","https://openalex.org/W2951100320"],"abstract_inverted_index":{"This":[0,82],"paper":[1],"proposes":[2],"a":[3,17,50,54,63,100],"new":[4],"regularization":[5,38,108],"algorithm":[6,83,119,148],"referred":[7],"to":[8,31,62,72,79],"as":[9],"macro-block":[10,51],"dropout.":[11],"The":[12,26],"overfitting":[13],"issue":[14],"has":[15,29,84,106],"been":[16],"difficult":[18],"problem":[19],"in":[20],"training":[21],"large":[22,55],"neural":[23],"network":[24],"models.":[25],"dropout":[27,71,78,103,135,161],"technique":[28],"proven":[30],"be":[32],"simple":[33],"yet":[34],"very":[35],"effective":[36],"for":[37,93],"by":[39],"preventing":[40],"complex":[41],"co-adaptations":[42],"during":[43],"training.":[44],"In":[45,110],"our":[46,111],"work,":[47],"we":[48,75,98],"define":[49],"that":[52],"contains":[53],"number":[56],"of":[57,87],"units":[58],"from":[59],"the":[60,85,133,159,163],"input":[61],"Recurrent":[64,114],"Neural":[65,115],"Network":[66],"(RNN).":[67],"Rather":[68],"than":[69],"applying":[70,88],"each":[73,80,94],"unit,":[74],"apply":[76],"random":[77],"macro-block.":[81],"effect":[86],"different":[89],"drop":[90],"out":[91],"rates":[92],"layer":[95],"even":[96],"if":[97],"keep":[99],"constant":[101],"average":[102],"rate,":[104],"which":[105],"better":[107],"effects.":[109],"experiments":[112],"using":[113],"Network-Transducer":[116],"(RNN-T),":[117],"this":[118,147],"shows":[120,149],"relatively":[121,150],"4.30":[122],"%":[123,126,152,155],"and":[124,139,153],"6.13":[125],"Word":[127],"Error":[128],"Rates":[129],"(WERs)":[130],"improvement":[131,157],"over":[132,158],"conventional":[134,160],"on":[136,162],"LibriSpeech":[137],"test-clean":[138],"test-other.":[140],"With":[141],"an":[142],"Attention-based":[143],"Encoder-Decoder":[144],"(AED)":[145],"model,":[146],"4.36":[151],"5.85":[154],"WERs":[156],"same":[164],"test":[165],"sets.":[166]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}