{"id":"https://openalex.org/W7148244131","doi":"https://doi.org/10.1109/asru65441.2025.11434800","title":"A Self-Refining Framework for Enhancing ASR Using TTS-Synthesized Data","display_name":"A Self-Refining Framework for Enhancing ASR Using TTS-Synthesized Data","publication_year":2025,"publication_date":"2025-12-06","ids":{"openalex":"https://openalex.org/W7148244131","doi":"https://doi.org/10.1109/asru65441.2025.11434800"},"language":null,"primary_location":{"id":"doi:10.1109/asru65441.2025.11434800","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru65441.2025.11434800","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","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/A5108893915","display_name":"C.-Y. Chou","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Cheng-Kang Chou","raw_affiliation_strings":["MediaTek Research"],"affiliations":[{"raw_affiliation_string":"MediaTek Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059902893","display_name":"Chan-Jan Hsu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chan-Jan Hsu","raw_affiliation_strings":["MediaTek Research"],"affiliations":[{"raw_affiliation_string":"MediaTek Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111114748","display_name":"Ho-Lam Chung","orcid":"https://orcid.org/0000-0003-3847-0166"},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Ho-Lam Chung","raw_affiliation_strings":["National Taiwan University"],"affiliations":[{"raw_affiliation_string":"National Taiwan University","institution_ids":["https://openalex.org/I16733864"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Liang-Hsuan Tseng","orcid":null},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Liang-Hsuan Tseng","raw_affiliation_strings":["National Taiwan University"],"affiliations":[{"raw_affiliation_string":"National Taiwan University","institution_ids":["https://openalex.org/I16733864"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Hsi-Chun Cheng","orcid":null},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Hsi-Chun Cheng","raw_affiliation_strings":["National Taiwan University"],"affiliations":[{"raw_affiliation_string":"National Taiwan University","institution_ids":["https://openalex.org/I16733864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071751103","display_name":"Yu-Kuan Fu","orcid":null},"institutions":[{"id":"https://openalex.org/I1304085615","display_name":"Nvidia (United Kingdom)","ror":"https://ror.org/02kr42612","country_code":"GB","type":"company","lineage":["https://openalex.org/I1304085615","https://openalex.org/I4210127875"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yu-Kuan Fu","raw_affiliation_strings":["Nvidia"],"affiliations":[{"raw_affiliation_string":"Nvidia","institution_ids":["https://openalex.org/I1304085615"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Kuan Po Huang","orcid":null},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Kuan Po Huang","raw_affiliation_strings":["National Taiwan University"],"affiliations":[{"raw_affiliation_string":"National Taiwan University","institution_ids":["https://openalex.org/I16733864"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5132807525","display_name":"Hung-Yi Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Hung-Yi Lee","raw_affiliation_strings":["National Taiwan University"],"affiliations":[{"raw_affiliation_string":"National Taiwan University","institution_ids":["https://openalex.org/I16733864"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5108893915"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.86676896,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9429000020027161,"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":0.9429000020027161,"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":0.01720000058412552,"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/T10403","display_name":"Phonetics and Phonology Research","score":0.011900000274181366,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.6212000250816345},{"id":"https://openalex.org/keywords/mandarin-chinese","display_name":"Mandarin Chinese","score":0.45339998602867126},{"id":"https://openalex.org/keywords/labeled-data","display_name":"Labeled data","score":0.27079999446868896},{"id":"https://openalex.org/keywords/synthetic-data","display_name":"Synthetic data","score":0.26489999890327454}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7786999940872192},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.6212000250816345},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5728999972343445},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5385000109672546},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5274999737739563},{"id":"https://openalex.org/C138954614","wikidata":"https://www.wikidata.org/wiki/Q9192","display_name":"Mandarin Chinese","level":2,"score":0.45339998602867126},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.29350000619888306},{"id":"https://openalex.org/C2776145971","wikidata":"https://www.wikidata.org/wiki/Q30673951","display_name":"Labeled data","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.26489999890327454},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.263700008392334}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asru65441.2025.11434800","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru65441.2025.11434800","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322410","display_name":"MediaTek","ror":"https://ror.org/05g9jck81"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W2911876655","https://openalex.org/W2962699523","https://openalex.org/W2963796886","https://openalex.org/W3162244132","https://openalex.org/W4372260202","https://openalex.org/W4385822890","https://openalex.org/W4385823377","https://openalex.org/W4391021545","https://openalex.org/W4392903389","https://openalex.org/W4392903468","https://openalex.org/W4392904149","https://openalex.org/W4392931276","https://openalex.org/W4392931630","https://openalex.org/W4401609216","https://openalex.org/W4402111998","https://openalex.org/W4402671236","https://openalex.org/W4404783543","https://openalex.org/W4406461879","https://openalex.org/W4408345790","https://openalex.org/W4417338252"],"related_works":[],"abstract_inverted_index":{"We":[0,54],"propose":[1],"a":[2,32,70,87,117,126],"self-refining":[3],"framework":[4,60,115],"that":[5],"enhances":[6],"ASR":[7,19,47,131],"performance":[8,132],"with":[9,16],"only":[10],"unlabeled":[11,68],"datasets.":[12],"The":[13],"process":[14],"starts":[15],"an":[17],"existing":[18],"model":[20],"generating":[21],"pseudo-labels":[22],"on":[23,61,100,105],"unannotated":[24],"speech,":[25,69],"which":[26],"are":[27,42],"then":[28],"used":[29],"to":[30,97,110,120],"train":[31],"highfidelity":[33],"text-to-speech":[34],"(TTS)":[35],"system.":[36],"Then,":[37],"synthesized":[38],"speech":[39],"text":[40,74],"pairs":[41],"bootstrapped":[43],"into":[44,86],"the":[45,50,56,59,80,114],"original":[46],"system,":[48],"completing":[49],"closed-loop":[51],"self-improvement":[52],"cycle.":[53],"demonstrated":[55],"effectiveness":[57],"of":[58,67,73],"Taiwanese-Mandarin":[62],"speech.":[63],"Leveraging":[64],"6,000":[65],"hours":[66],"moderate":[71],"amount":[72],"data,":[75],"and":[76,102,124],"synthetic":[77],"content":[78],"from":[79],"AI":[81],"models,":[82],"we":[83],"adapt":[84],"Whisper-large-v2":[85],"specialized":[88],"model,":[89],"Twister.":[90],"Twister":[91],"reduces":[92],"error":[93],"rates":[94],"by":[95],"up":[96],"$20":[98],"\\%$":[99,104],"Mandarin":[101],"$50":[103],"Mandarin-English":[106],"code-switching":[107],"benchmarks":[108],"compared":[109],"Whisper.":[111],"Results":[112],"highlight":[113],"as":[116],"compelling":[118],"alternative":[119],"pseudo-labeling":[121],"self-distillation":[122],"approaches":[123],"provides":[125],"practical":[127],"pathway":[128],"for":[129],"improving":[130],"in":[133],"lowresource":[134],"or":[135],"domain-specific":[136],"settings.":[137]},"counts_by_year":[],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2026-04-03T00:00:00"}