{"id":"https://openalex.org/W4372259982","doi":"https://doi.org/10.1109/icassp49357.2023.10096146","title":"Repackagingaugment: Overcoming Prediction Error Amplification in Weight-Averaged Speech Recognition Models Subject to Self-Training","display_name":"Repackagingaugment: Overcoming Prediction Error Amplification in Weight-Averaged Speech Recognition Models Subject to Self-Training","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372259982","doi":"https://doi.org/10.1109/icassp49357.2023.10096146"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10096146","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp49357.2023.10096146","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":"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/A5100705624","display_name":"Jaehong Lee","orcid":"https://orcid.org/0000-0002-5056-829X"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jae-Hong Lee","raw_affiliation_strings":["Hanyang University,Department of Electronics Engineering,Seoul,Republic of Korea","Department of Electronics Engineering, Hanyang University, Seoul, Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hanyang University,Department of Electronics Engineering,Seoul,Republic of Korea","institution_ids":["https://openalex.org/I4575257"]},{"raw_affiliation_string":"Department of Electronics Engineering, Hanyang University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100358143","display_name":"Donghyun Kim","orcid":"https://orcid.org/0000-0002-6626-3706"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Dong-Hyun Kim","raw_affiliation_strings":["Hanyang University,Department of Electronics Engineering,Seoul,Republic of Korea","Department of Electronics Engineering, Hanyang University, Seoul, Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hanyang University,Department of Electronics Engineering,Seoul,Republic of Korea","institution_ids":["https://openalex.org/I4575257"]},{"raw_affiliation_string":"Department of Electronics Engineering, Hanyang University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002418613","display_name":"Joon\u2010Hyuk Chang","orcid":"https://orcid.org/0000-0003-2610-2323"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Joon-Hyuk Chang","raw_affiliation_strings":["Hanyang University,Department of Electronics Engineering,Seoul,Republic of Korea","Department of Electronics Engineering, Hanyang University, Seoul, Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hanyang University,Department of Electronics Engineering,Seoul,Republic of Korea","institution_ids":["https://openalex.org/I4575257"]},{"raw_affiliation_string":"Department of Electronics Engineering, Hanyang University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I4575257"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4575257"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"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":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/T10181","display_name":"Natural Language Processing Techniques","score":0.9987999796867371,"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":0.9983999729156494,"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/computer-science","display_name":"Computer science","score":0.7666553258895874},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.6536014080047607},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.6001600027084351},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.5615900754928589},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5516678094863892},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4939592480659485},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.43166911602020264},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3968995213508606},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.35285788774490356},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1404183804988861}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7666553258895874},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.6536014080047607},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.6001600027084351},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.5615900754928589},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5516678094863892},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4939592480659485},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.43166911602020264},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3968995213508606},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.35285788774490356},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1404183804988861},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","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},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0},{"id":"https://openalex.org/C94625758","wikidata":"https://www.wikidata.org/wiki/Q7163","display_name":"Politics","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49357.2023.10096146","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp49357.2023.10096146","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":[{"score":0.44999998807907104,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1922655562","https://openalex.org/W2127141656","https://openalex.org/W2129838208","https://openalex.org/W2407080277","https://openalex.org/W2617258110","https://openalex.org/W2667408400","https://openalex.org/W2788838181","https://openalex.org/W2933138175","https://openalex.org/W2936774411","https://openalex.org/W2953070460","https://openalex.org/W2953190524","https://openalex.org/W2963173418","https://openalex.org/W2973049979","https://openalex.org/W2979476256","https://openalex.org/W3015522062","https://openalex.org/W3026041220","https://openalex.org/W3036601975","https://openalex.org/W3043783436","https://openalex.org/W3093579165","https://openalex.org/W3096338464","https://openalex.org/W3160525311","https://openalex.org/W3196824004","https://openalex.org/W3209059054","https://openalex.org/W4221145109","https://openalex.org/W4221155125","https://openalex.org/W4226033575","https://openalex.org/W4281251058","https://openalex.org/W4311140058","https://openalex.org/W4322503680","https://openalex.org/W6640090968","https://openalex.org/W6733814495","https://openalex.org/W6748173244","https://openalex.org/W6749107692","https://openalex.org/W6769196770","https://openalex.org/W6780218876","https://openalex.org/W6784614252","https://openalex.org/W6810007534","https://openalex.org/W6810127701","https://openalex.org/W6810933673"],"related_works":["https://openalex.org/W2062195135","https://openalex.org/W1569283511","https://openalex.org/W2795079307","https://openalex.org/W2793058541","https://openalex.org/W1983629434","https://openalex.org/W4236193183","https://openalex.org/W2053866214","https://openalex.org/W2607505004","https://openalex.org/W2055929693","https://openalex.org/W2296205523"],"abstract_inverted_index":{"Representation-based":[0],"speech":[1],"recognition":[2],"models":[3,12,96,135],"have":[4,57],"demonstrated":[5],"state-of-the-art":[6],"performance":[7,168],"on":[8,15,20,152],"downstream":[9],"tasks.":[10],"These":[11],"are":[13],"pre-trained":[14],"large-scale":[16],"unlabeled":[17],"data,":[18,26],"fine-tuned":[19],"a":[21,38,65,87,124],"small":[22],"amount":[23],"of":[24,67,76,95,128,145,169],"labeled":[25],"and":[27,120,159],"subsequently":[28],"advanced":[29],"via":[30],"the":[31,52,62,73,93,99,113,134,143,167],"self-training":[32],"procedure":[33],"by":[34,46,141],"leveraging":[35],"pseudo-labels.":[36],"However,":[37],"self-trained":[39,78],"representation":[40,153],"model":[41],"produces":[42],"prediction":[43,74,139],"errors":[44,75,140],"caused":[45],"training":[47],"with":[48],"incorrect":[49,101,146],"labels":[50,102],"in":[51,64,106],"pseudo-labeled":[53],"data.":[54],"Weight-averaging":[55],"methods":[56,71],"been":[58],"employed":[59],"to":[60,136],"refine":[61],"pseudo-labels":[63],"variety":[66],"studies;":[68],"however,":[69],"these":[70],"amplify":[72],"each":[77],"model.":[79],"To":[80],"alleviate":[81],"this":[82],"problem,":[83],"we":[84,161],"propose":[85],"RepackagingAugment,":[86],"data":[88,110,116],"augmentation":[89,111],"method":[90],"that":[91,163],"improves":[92,166],"diversity":[94],"while":[97],"preventing":[98],"same":[100],"from":[103],"recursively":[104],"occurring":[105],"every":[107],"epoch.":[108],"Our":[109],"deconstructs":[112],"paired":[114],"speech\u2013text":[115],"into":[117,123],"word":[118,129],"units":[119],"repackages":[121],"them":[122],"randomly":[125],"determined":[126],"number":[127],"sequences.":[130],"This":[131],"strategy":[132],"induces":[133],"produce":[137],"different":[138],"mitigating":[142],"problem":[144],"label":[147],"over-fitting.":[148],"Through":[149],"various":[150],"experiments":[151],"models,":[154],"such":[155],"as":[156],"wav2vec":[157],"2.0":[158],"data2vec,":[160],"demonstrate":[162],"our":[164],"approach":[165],"weight-averaged":[170],"models.":[171]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}