{"id":"https://openalex.org/W4408352078","doi":"https://doi.org/10.1109/icassp49660.2025.10887642","title":"Large Language Model Can Transcribe Speech in Multi-Talker Scenarios with Versatile Instructions","display_name":"Large Language Model Can Transcribe Speech in Multi-Talker Scenarios with Versatile Instructions","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408352078","doi":"https://doi.org/10.1109/icassp49660.2025.10887642"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10887642","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10887642","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/A5031382920","display_name":"Lingwei Meng","orcid":null},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Lingwei Meng","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026334377","display_name":"Shujie Hu","orcid":"https://orcid.org/0000-0002-8475-4912"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Shujie Hu","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062761975","display_name":"Jiawen Kang","orcid":"https://orcid.org/0000-0002-8218-3490"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Jiawen Kang","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101675943","display_name":"Zhaoqing Li","orcid":"https://orcid.org/0009-0002-5504-7802"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zhaoqing Li","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061240757","display_name":"Yuejiao Wang","orcid":"https://orcid.org/0000-0002-8221-8841"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yuejiao Wang","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101759485","display_name":"Wenxuan Wu","orcid":"https://orcid.org/0000-0002-5371-1186"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Wenxuan Wu","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025119324","display_name":"Xixin Wu","orcid":"https://orcid.org/0000-0001-9543-1572"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xixin Wu","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037109470","display_name":"Xunying Liu","orcid":"https://orcid.org/0000-0001-6725-1160"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xunying Liu","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5019458385","display_name":"Helen Meng","orcid":"https://orcid.org/0000-0002-4427-3532"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Helen Meng","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I177725633"],"apc_list":null,"apc_paid":null,"fwci":11.7111,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.98499415,"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":0.9517999887466431,"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":0.9517999887466431,"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/computer-science","display_name":"Computer science","score":0.8156555891036987},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5632334351539612},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.47759437561035156},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4615474045276642}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8156555891036987},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5632334351539612},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.47759437561035156},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4615474045276642}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10887642","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10887642","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":[{"display_name":"Quality Education","score":0.5400000214576721,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2460742184","https://openalex.org/W3013139777","https://openalex.org/W3015834770","https://openalex.org/W3017474798","https://openalex.org/W3035390927","https://openalex.org/W3186652349","https://openalex.org/W3196194966","https://openalex.org/W3196509775","https://openalex.org/W3209984917","https://openalex.org/W4226491018","https://openalex.org/W4372259971","https://openalex.org/W4372342485","https://openalex.org/W4372349800","https://openalex.org/W4385807453","https://openalex.org/W4385822384","https://openalex.org/W4385823037","https://openalex.org/W4387010481","https://openalex.org/W4392902710","https://openalex.org/W4402111585","https://openalex.org/W4402112082","https://openalex.org/W4402116591","https://openalex.org/W4403640427","https://openalex.org/W4403863362","https://openalex.org/W4404781855","https://openalex.org/W4408352376","https://openalex.org/W6759579507","https://openalex.org/W6796581206","https://openalex.org/W6847363464","https://openalex.org/W6850015000","https://openalex.org/W6857054612","https://openalex.org/W6857351970"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Recent":[0],"advancements":[1],"in":[2,19,28,45,48,131,148],"large":[3],"language":[4],"models":[5],"(LLMs)":[6],"have":[7,23],"revolutionized":[8],"various":[9],"domains,":[10],"bringing":[11],"significant":[12],"progress":[13,18],"and":[14,64,77,84,98,118],"new":[15],"opportunities.":[16],"Despite":[17],"speech-related":[20,142],"tasks,":[21],"LLMs":[22,44],"not":[24],"been":[25],"sufficiently":[26],"explored":[27],"multi-talker":[29,49,56],"scenarios.":[30],"In":[31],"this":[32],"work,":[33],"we":[34],"present":[35],"a":[36],"pioneering":[37],"effort":[38],"to":[39,55,87,95,140],"investigate":[40],"the":[41,113,123,136],"capability":[42],"of":[43,126,138],"transcribing":[46],"speech":[47,58,90,116],"environments,":[50],"following":[51],"versatile":[52],"instructions":[53,147],"related":[54],"automatic":[57],"recognition":[59],"(ASR),":[60],"target":[61],"talker":[62,69],"ASR,":[63],"ASR":[65],"based":[66,144],"on":[67,145],"specific":[68],"attributes":[70],"such":[71,149],"as":[72],"sex,":[73],"occurrence":[74],"order,":[75],"language,":[76],"keyword":[78],"spoken.":[79],"Our":[80],"approach":[81],"utilizes":[82],"WavLM":[83],"Whisper":[85],"encoder":[86],"extract":[88],"multi-faceted":[89],"representations":[91,102],"that":[92],"are":[93,103],"sensitive":[94],"speaker":[96],"characteristics":[97],"semantic":[99],"context.":[100],"These":[101],"then":[104],"fed":[105],"into":[106],"an":[107],"LLM":[108,139],"fine-tuned":[109],"using":[110],"LoRA,":[111],"enabling":[112],"capabilities":[114],"for":[115],"comprehension":[117],"transcription.":[119],"Comprehensive":[120],"experiments":[121],"reveal":[122],"promising":[124],"performance":[125],"our":[127],"proposed":[128],"system,":[129],"MT-LLM,":[130],"cocktail":[132],"party":[133],"scenarios,":[134],"highlighting":[135],"potential":[137],"handle":[141],"tasks":[143],"user":[146],"complex":[150],"settings1.":[151]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":5}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}