{"id":"https://openalex.org/W4319862660","doi":"https://doi.org/10.1109/slt54892.2023.10022760","title":"Interdecoder: using Attention Decoders as Intermediate Regularization for CTC-Based Speech Recognition","display_name":"Interdecoder: using Attention Decoders as Intermediate Regularization for CTC-Based Speech Recognition","publication_year":2023,"publication_date":"2023-01-09","ids":{"openalex":"https://openalex.org/W4319862660","doi":"https://doi.org/10.1109/slt54892.2023.10022760"},"language":"en","primary_location":{"id":"doi:10.1109/slt54892.2023.10022760","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/slt54892.2023.10022760","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/A5113730169","display_name":"Tatsuya Komatsu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210096607","display_name":"Line Corporation (Japan)","ror":"https://ror.org/00qg8pm87","country_code":"JP","type":"company","lineage":["https://openalex.org/I4210096607","https://openalex.org/I60922564"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Tatsuya Komatsu","raw_affiliation_strings":["LINE Corporation,Tokyo,Japan","LINE Corporation, Tokyo, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"LINE Corporation,Tokyo,Japan","institution_ids":["https://openalex.org/I4210096607"]},{"raw_affiliation_string":"LINE Corporation, Tokyo, Japan","institution_ids":["https://openalex.org/I4210096607"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5044818016","display_name":"Yusuke Fujita","orcid":"https://orcid.org/0000-0002-6523-8146"},"institutions":[{"id":"https://openalex.org/I4210096607","display_name":"Line Corporation (Japan)","ror":"https://ror.org/00qg8pm87","country_code":"JP","type":"company","lineage":["https://openalex.org/I4210096607","https://openalex.org/I60922564"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Yusuke Fujita","raw_affiliation_strings":["LINE Corporation,Tokyo,Japan","LINE Corporation, Tokyo, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"LINE Corporation,Tokyo,Japan","institution_ids":["https://openalex.org/I4210096607"]},{"raw_affiliation_string":"LINE Corporation, Tokyo, Japan","institution_ids":["https://openalex.org/I4210096607"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210096607"],"apc_list":null,"apc_paid":null,"fwci":0.4356,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.54382729,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":94},"biblio":{"volume":null,"issue":null,"first_page":"46","last_page":"51"},"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":0.9980999827384949,"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":0.9976999759674072,"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/autoregressive-model","display_name":"Autoregressive model","score":0.817630410194397},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.7607938647270203},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7548724412918091},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6841737627983093},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6122003197669983},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5947245955467224},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.548547625541687},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5286417603492737},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.5193641781806946},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.41974180936813354},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.37655529379844666},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.2855077385902405},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.13264012336730957},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.09025639295578003},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08115619421005249}],"concepts":[{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.817630410194397},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7607938647270203},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7548724412918091},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6841737627983093},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6122003197669983},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5947245955467224},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.548547625541687},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5286417603492737},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5193641781806946},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41974180936813354},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.37655529379844666},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2855077385902405},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.13264012336730957},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.09025639295578003},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08115619421005249},{"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/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/slt54892.2023.10022760","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/slt54892.2023.10022760","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":[{"score":0.5299999713897705,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":29,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2251321385","https://openalex.org/W2407080277","https://openalex.org/W2766219058","https://openalex.org/W2892009249","https://openalex.org/W2936774411","https://openalex.org/W2962780374","https://openalex.org/W2963303028","https://openalex.org/W2963979492","https://openalex.org/W2972818416","https://openalex.org/W3007328579","https://openalex.org/W3015960524","https://openalex.org/W3097777922","https://openalex.org/W3097882114","https://openalex.org/W3112157188","https://openalex.org/W3169714379","https://openalex.org/W3197140813","https://openalex.org/W3206573929","https://openalex.org/W4226212269","https://openalex.org/W4385245566","https://openalex.org/W6623517193","https://openalex.org/W6631190155","https://openalex.org/W6631362777","https://openalex.org/W6638749077","https://openalex.org/W6687566353","https://openalex.org/W6691770337","https://openalex.org/W6774835902","https://openalex.org/W6784531339","https://openalex.org/W7027429494"],"related_works":["https://openalex.org/W2947912915","https://openalex.org/W2917049097","https://openalex.org/W3209239055","https://openalex.org/W4307320830","https://openalex.org/W4287775347","https://openalex.org/W3116857254","https://openalex.org/W3116268265","https://openalex.org/W4221144390","https://openalex.org/W2989134874","https://openalex.org/W4285757700"],"abstract_inverted_index":{"We":[0],"propose":[1],"InterDecoder:":[2],"a":[3,118],"new":[4],"non-autoregressive":[5,24],"automatic":[6],"speech":[7],"recognition":[8],"(NAR-ASR)":[9],"training":[10,51],"method":[11,116],"that":[12,113],"injects":[13],"the":[14,22,58,62,71,78,83,93,99,107,114,124,128],"advantage":[15],"of":[16],"token-wise":[17],"autoregressive":[18,34],"decoders":[19],"while":[20],"keeping":[21],"efficient":[23],"inference.":[25],"The":[26,48],"NAR-ASR":[27,130],"models":[28,35],"are":[29],"often":[30],"less":[31],"accurate":[32],"than":[33],"such":[36],"as":[37],"Transformer":[38],"decoder,":[39],"which":[40,91],"predict":[41],"tokens":[42],"conditioned":[43],"on":[44,106],"previously":[45],"predicted":[46],"tokens.":[47],"Inter-Decoder":[49],"regularizes":[50],"by":[52],"feeding":[53],"intermediate":[54],"encoder":[55],"outputs":[56],"into":[57],"decoder":[59,79],"to":[60,97,127],"compute":[61],"token-level":[63],"prediction":[64],"errors":[65],"given":[66],"previous":[67],"ground-truth":[68],"tokens,":[69],"whereas":[70],"widely":[72],"used":[73],"Hybrid":[74],"CTC/Attention":[75],"model":[76],"uses":[77,92],"loss":[80],"only":[81],"at":[82,123],"final":[84],"layer.":[85],"In":[86],"combination":[87],"with":[88],"Self-conditioned":[89],"CTC,":[90],"Intermediate":[94],"CTC":[95],"predictions":[96],"condition":[98],"encoder,":[100],"performance":[101],"is":[102],"further":[103],"improved.":[104],"Experiments":[105],"Librispeech":[108],"and":[109],"Tedlium2":[110],"dataset":[111],"show":[112],"proposed":[115],"shows":[117],"relative":[119],"6%":[120],"WER":[121],"improvement":[122],"maximum":[125],"compared":[126],"conventional":[129],"methods.":[131]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}