{"id":"https://openalex.org/W4408345694","doi":"https://doi.org/10.1109/icassp49660.2025.10888227","title":"JELLY: Joint Emotion Recognition and Context Reasoning with LLMs for Conversational Speech Synthesis","display_name":"JELLY: Joint Emotion Recognition and Context Reasoning with LLMs for Conversational Speech Synthesis","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408345694","doi":"https://doi.org/10.1109/icassp49660.2025.10888227"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10888227","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888227","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":null,"display_name":"Jun-Hyeok Cha","orcid":null},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jun-Hyeok Cha","raw_affiliation_strings":["Korea University,Dept. of Artificial Intelligence,Seoul,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Dept. of Artificial Intelligence,Seoul,Korea","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007493723","display_name":"Seung-Bin Kim","orcid":"https://orcid.org/0000-0002-2287-9111"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seung-Bin Kim","raw_affiliation_strings":["Korea University,Dept. of Artificial Intelligence,Seoul,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Dept. of Artificial Intelligence,Seoul,Korea","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111861907","display_name":"Hyung\u2010Seok Oh","orcid":null},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyung-Seok Oh","raw_affiliation_strings":["Korea University,Dept. of Artificial Intelligence,Seoul,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Dept. of Artificial Intelligence,Seoul,Korea","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5064160666","display_name":"Seong-Whan Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seong-Whan Lee","raw_affiliation_strings":["Korea University,Dept. of Artificial Intelligence,Seoul,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Dept. of Artificial Intelligence,Seoul,Korea","institution_ids":["https://openalex.org/I197347611"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I197347611"],"apc_list":null,"apc_paid":null,"fwci":3.346,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.90263792,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":98},"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":0.7019000053405762,"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":0.7019000053405762,"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/joint","display_name":"Joint (building)","score":0.6757184267044067},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6402403116226196},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.631773829460144},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4689981937408447},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.41604992747306824},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.35306328535079956},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.3341952860355377},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.09187692403793335},{"id":"https://openalex.org/keywords/history","display_name":"History","score":0.08460432291030884}],"concepts":[{"id":"https://openalex.org/C18555067","wikidata":"https://www.wikidata.org/wiki/Q8375051","display_name":"Joint (building)","level":2,"score":0.6757184267044067},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6402403116226196},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.631773829460144},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4689981937408447},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.41604992747306824},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35306328535079956},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3341952860355377},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.09187692403793335},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.08460432291030884},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.0},{"id":"https://openalex.org/C170154142","wikidata":"https://www.wikidata.org/wiki/Q150737","display_name":"Architectural engineering","level":1,"score":0.0},{"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/icassp49660.2025.10888227","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888227","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":[{"id":"https://openalex.org/F4320321373","display_name":"Korea University","ror":"https://ror.org/047dqcg40"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":39,"referenced_works":["https://openalex.org/W2030931454","https://openalex.org/W2083677758","https://openalex.org/W2146334809","https://openalex.org/W2897353073","https://openalex.org/W2964243274","https://openalex.org/W2989568715","https://openalex.org/W2989793251","https://openalex.org/W3082414409","https://openalex.org/W3088811935","https://openalex.org/W3099284785","https://openalex.org/W3151309757","https://openalex.org/W3168867926","https://openalex.org/W4224926192","https://openalex.org/W4225300652","https://openalex.org/W4372260474","https://openalex.org/W4375868859","https://openalex.org/W4385807453","https://openalex.org/W4391021627","https://openalex.org/W4392902952","https://openalex.org/W4392904491","https://openalex.org/W4392904607","https://openalex.org/W4392931276","https://openalex.org/W4393160807","https://openalex.org/W4396542467","https://openalex.org/W4402115961","https://openalex.org/W4402669711","https://openalex.org/W6745297980","https://openalex.org/W6753049143","https://openalex.org/W6757817989","https://openalex.org/W6778823374","https://openalex.org/W6780218876","https://openalex.org/W6783867762","https://openalex.org/W6847363464","https://openalex.org/W6849177959","https://openalex.org/W6850625674","https://openalex.org/W6860346458","https://openalex.org/W6861503643","https://openalex.org/W6869881118","https://openalex.org/W6872347754"],"related_works":["https://openalex.org/W1996130883","https://openalex.org/W2748574964","https://openalex.org/W2888483922","https://openalex.org/W4396737233","https://openalex.org/W2367747139","https://openalex.org/W4391102217","https://openalex.org/W2566187525","https://openalex.org/W2566334511","https://openalex.org/W2367150592","https://openalex.org/W2378889330"],"abstract_inverted_index":{"Recently,":[0],"there":[1],"has":[2],"been":[3],"a":[4,28,47],"growing":[5],"demand":[6],"for":[7,39,101],"conversational":[8,20,94,132],"speech":[9,16,42,78,95,105,120,133],"synthesis":[10],"(CSS)":[11],"that":[12,32,112,121],"generates":[13],"more":[14],"natural":[15],"by":[17,45],"considering":[18],"the":[19,65,128],"context.":[21],"To":[22],"address":[23],"this,":[24],"we":[25],"introduce":[26],"JELLY,":[27],"novel":[29],"CSS":[30],"framework":[31],"integrates":[33],"emotion":[34],"recognition":[35],"and":[36],"context":[37,100,117],"reasoning":[38],"generating":[40,102],"appropriate":[41,104],"in":[43,70,106,115],"conversation":[44],"fine-tuning":[46],"large":[48],"language":[49],"model":[50,89],"(LLM)":[51],"with":[52,80,93,124],"multiple":[53],"partial":[54],"LoRA":[55],"modules.":[56],"We":[57],"propose":[58],"an":[59],"Emotion-aware":[60],"Q-former":[61],"encoder,":[62],"which":[63],"enables":[64],"LLM":[66],"to":[67,76,97],"perceive":[68],"emotions":[69,79],"speech.":[71,86],"The":[72,87],"encoder":[73],"is":[74,90],"trained":[75],"align":[77],"text,":[81],"utilizing":[82],"datasets":[83],"of":[84,130],"emotional":[85,99,116,131],"entire":[88],"then":[91],"fine-tuned":[92],"data":[96],"infer":[98],"emotionally":[103],"conversation.":[107],"Our":[108],"experimental":[109],"results":[110],"demonstrate":[111],"JELLY":[113],"excels":[114],"modeling,":[118],"synthesizing":[119],"naturally":[122],"aligns":[123],"conversation,":[125],"while":[126],"mitigating":[127],"scarcity":[129],"datasets.":[134]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}