{"id":"https://openalex.org/W4408346038","doi":"https://doi.org/10.1109/icassp49660.2025.10890030","title":"Boosting Code-Switching ASR with Mixture of Experts Enhanced Speech-Conditioned LLM","display_name":"Boosting Code-Switching ASR with Mixture of Experts Enhanced Speech-Conditioned LLM","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408346038","doi":"https://doi.org/10.1109/icassp49660.2025.10890030"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10890030","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890030","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 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/A5027555657","display_name":"Fengrun Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fengrun Zhang","raw_affiliation_strings":["Kuaishou Technology,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kuaishou Technology,Beijing,China","institution_ids":["https://openalex.org/I4401726859"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114246682","display_name":"Geng Wang","orcid":"https://orcid.org/0009-0004-2832-4207"},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang Geng","raw_affiliation_strings":["Kuaishou Technology,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kuaishou Technology,Beijing,China","institution_ids":["https://openalex.org/I4401726859"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109692702","display_name":"Hukai Huang","orcid":null},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hukai Huang","raw_affiliation_strings":["Kuaishou Technology,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kuaishou Technology,Beijing,China","institution_ids":["https://openalex.org/I4401726859"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022676888","display_name":"Yahui Shan","orcid":"https://orcid.org/0000-0002-1510-2590"},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yahui Shan","raw_affiliation_strings":["Kuaishou Technology,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kuaishou Technology,Beijing,China","institution_ids":["https://openalex.org/I4401726859"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070207133","display_name":"Yi Cheng","orcid":"https://orcid.org/0000-0002-5963-5751"},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cheng Yi","raw_affiliation_strings":["Kuaishou Technology,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kuaishou Technology,Beijing,China","institution_ids":["https://openalex.org/I4401726859"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101319176","display_name":"He Qu","orcid":null},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"He Qu","raw_affiliation_strings":["Kuaishou Technology,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kuaishou Technology,Beijing,China","institution_ids":["https://openalex.org/I4401726859"]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4401726859"],"apc_list":null,"apc_paid":null,"fwci":5.019,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.94326755,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":97},"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":0.9965000152587891,"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":0.9965000152587891,"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/T12031","display_name":"Speech and dialogue systems","score":0.9959999918937683,"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":0.9835000038146973,"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/boosting","display_name":"Boosting (machine learning)","score":0.8432128429412842},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7251852750778198},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5913437008857727},{"id":"https://openalex.org/keywords/code-switching","display_name":"Code-switching","score":0.5642621517181396},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.40653520822525024},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3845195770263672},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.07319003343582153}],"concepts":[{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.8432128429412842},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7251852750778198},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5913437008857727},{"id":"https://openalex.org/C18552078","wikidata":"https://www.wikidata.org/wiki/Q255615","display_name":"Code-switching","level":2,"score":0.5642621517181396},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.40653520822525024},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3845195770263672},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.07319003343582153},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10890030","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890030","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2892009249","https://openalex.org/W2963242190","https://openalex.org/W3096122506","https://openalex.org/W3097777922","https://openalex.org/W3200947677","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4391021666","https://openalex.org/W4391021741","https://openalex.org/W4400434308","https://openalex.org/W4402112167","https://openalex.org/W4402112425","https://openalex.org/W4405709396","https://openalex.org/W4406461663","https://openalex.org/W6739901393","https://openalex.org/W6780815891","https://openalex.org/W6796581206","https://openalex.org/W6847363464","https://openalex.org/W6851775633","https://openalex.org/W6857054612","https://openalex.org/W6857062747","https://openalex.org/W6859099255","https://openalex.org/W6870165730","https://openalex.org/W6870775139","https://openalex.org/W6871553294"],"related_works":["https://openalex.org/W2771594921","https://openalex.org/W2432897346","https://openalex.org/W2181336723","https://openalex.org/W4389976243","https://openalex.org/W3158134258","https://openalex.org/W2125652721","https://openalex.org/W3138119129","https://openalex.org/W1540371141","https://openalex.org/W2940588741","https://openalex.org/W3204019825"],"abstract_inverted_index":{"In":[0],"this":[1],"paper,":[2],"we":[3,34,85],"introduce":[4],"a":[5,13,60,87],"speech-conditioned":[6],"Large":[7],"Language":[8],"Model":[9],"(LLM)":[10],"integrated":[11],"with":[12,62,99,121],"Mixture":[14],"of":[15,24,40,51,75,83],"Experts":[16],"(MoE)":[17],"based":[18],"connector":[19,61,94,112],"to":[20,53,102,106,131],"address":[21],"the":[22,73,80,107,122],"challenge":[23],"Code-Switching":[25],"(CS)":[26],"scenario":[27],"in":[28],"Automatic":[29],"Speech":[30],"Recognition":[31],"(ASR).":[32],"Specifically,":[33],"propose":[35,86],"an":[36],"Insertion":[37],"and":[38,78,97,113,126,147],"Deletion":[39],"Interruption":[41],"Token":[42],"(IDIT)":[43],"mechanism":[44,125],"for":[45],"better":[46],"transfer":[47],"text":[48,108],"generation":[49],"ability":[50],"LLM":[52,114],"speech":[54,104],"recognition":[55],"task.":[56],"We":[57],"also":[58],"present":[59],"MoE":[63],"architecture":[64],"that":[65,138],"manages":[66],"multiple":[67,76],"languages":[68],"efficiently.":[69],"To":[70],"further":[71],"enhance":[72],"collaboration":[74],"experts":[77,101,128],"leverage":[79],"understanding":[81],"capabilities":[82],"LLM,":[84],"two-stage":[88],"progressive":[89],"training":[90],"strategy:":[91],"1)":[92],"The":[93,111],"is":[95],"unfrozen":[96],"trained":[98,120],"language-specialized":[100],"map":[103],"representations":[105],"space.":[109],"2)":[110],"Lower-Rank":[115],"Adaptation":[116],"(LoRA)":[117],"adaptor":[118],"are":[119,129],"proposed":[123],"IDIT":[124],"all":[127],"activated":[130],"learn":[132],"general":[133],"representations.":[134],"Experimental":[135],"results":[136],"demonstrate":[137],"our":[139],"method":[140],"significantly":[141],"outperforms":[142],"state-of-the-art":[143],"models,":[144],"including":[145],"end-to-end":[146],"large-scale":[148],"audio-language":[149],"models.":[150]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}