{"id":"https://openalex.org/W4408351812","doi":"https://doi.org/10.1109/icassp49660.2025.10890141","title":"Get Large Language Models Ready to Speak: A Late-fusion Approach for Speech Generation","display_name":"Get Large Language Models Ready to Speak: A Late-fusion Approach for Speech Generation","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408351812","doi":"https://doi.org/10.1109/icassp49660.2025.10890141"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10890141","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890141","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":"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":null,"display_name":"Maohao Shen","orcid":null},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Maohao Shen","raw_affiliation_strings":["Massachusetts Institute of Technology"],"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021863803","display_name":"Shun Zhang","orcid":"https://orcid.org/0000-0002-3493-1247"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shun Zhang","raw_affiliation_strings":["AI at Meta"],"affiliations":[{"raw_affiliation_string":"AI at Meta","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057239023","display_name":"Jilong Wu","orcid":"https://orcid.org/0009-0007-8000-347X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jilong Wu","raw_affiliation_strings":["AI at Meta"],"affiliations":[{"raw_affiliation_string":"AI at Meta","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085174046","display_name":"Zhiping Xiu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhiping Xiu","raw_affiliation_strings":["AI at Meta"],"affiliations":[{"raw_affiliation_string":"AI at Meta","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076974550","display_name":"Ehab A. AlBadawy","orcid":"https://orcid.org/0000-0003-3954-733X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ehab AlBadawy","raw_affiliation_strings":["AI at Meta"],"affiliations":[{"raw_affiliation_string":"AI at Meta","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102804529","display_name":"Yiting Lu","orcid":"https://orcid.org/0009-0003-0807-9972"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yiting Lu","raw_affiliation_strings":["AI at Meta"],"affiliations":[{"raw_affiliation_string":"AI at Meta","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113773386","display_name":"Mike Seltzer","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mike Seltzer","raw_affiliation_strings":["AI at Meta"],"affiliations":[{"raw_affiliation_string":"AI at Meta","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100734672","display_name":"Qing He","orcid":"https://orcid.org/0000-0001-8833-5398"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qing He","raw_affiliation_strings":["AI at Meta"],"affiliations":[{"raw_affiliation_string":"AI at Meta","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":8,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I63966007"],"apc_list":null,"apc_paid":null,"fwci":4.4566,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.93511954,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":96},"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":0.9836000204086304,"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":0.9836000204086304,"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/T10028","display_name":"Topic Modeling","score":0.9298999905586243,"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":0.9247000217437744,"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":0.7379180192947388},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5037021040916443},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.49065056443214417},{"id":"https://openalex.org/keywords/fusion","display_name":"Fusion","score":0.4777851998806},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.43882066011428833},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.40814727544784546},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.3460402488708496},{"id":"https://openalex.org/keywords/philosophy","display_name":"Philosophy","score":0.05563068389892578}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7379180192947388},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5037021040916443},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.49065056443214417},{"id":"https://openalex.org/C158525013","wikidata":"https://www.wikidata.org/wiki/Q2593739","display_name":"Fusion","level":2,"score":0.4777851998806},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.43882066011428833},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.40814727544784546},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3460402488708496},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.05563068389892578}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10890141","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890141","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":38,"referenced_works":["https://openalex.org/W2903739847","https://openalex.org/W2963159690","https://openalex.org/W2964243274","https://openalex.org/W4307323391","https://openalex.org/W4381786045","https://openalex.org/W4387595589","https://openalex.org/W4389518827","https://openalex.org/W4389524500","https://openalex.org/W4390075359","https://openalex.org/W4391021666","https://openalex.org/W4392903704","https://openalex.org/W4392903956","https://openalex.org/W4392904805","https://openalex.org/W4398777692","https://openalex.org/W4402672020","https://openalex.org/W4408353629","https://openalex.org/W6749838110","https://openalex.org/W6763832098","https://openalex.org/W6771713106","https://openalex.org/W6782879696","https://openalex.org/W6790356757","https://openalex.org/W6796581206","https://openalex.org/W6805710207","https://openalex.org/W6810673746","https://openalex.org/W6848735303","https://openalex.org/W6850218400","https://openalex.org/W6850334629","https://openalex.org/W6852781825","https://openalex.org/W6852975727","https://openalex.org/W6853096648","https://openalex.org/W6853530687","https://openalex.org/W6853611000","https://openalex.org/W6853998256","https://openalex.org/W6856794988","https://openalex.org/W6857054612","https://openalex.org/W6861618871","https://openalex.org/W6868761718","https://openalex.org/W7073593060"],"related_works":["https://openalex.org/W2099421762","https://openalex.org/W2530546662","https://openalex.org/W2967030268","https://openalex.org/W2169518243","https://openalex.org/W2185253430","https://openalex.org/W4210345652","https://openalex.org/W1984333081","https://openalex.org/W2945063165","https://openalex.org/W2205042400","https://openalex.org/W2090790166"],"abstract_inverted_index":{"Large":[0],"language":[1,7],"models":[2],"(LLMs)":[3],"have":[4],"revolutionized":[5],"natural":[6],"processing":[8],"(NLP)":[9],"with":[10,24],"impressive":[11],"performance":[12,81],"across":[13],"various":[14],"text-based":[15],"tasks.":[16],"However,":[17],"the":[18],"extension":[19],"of":[20,116],"text-dominant":[21],"LLMs":[22],"to":[23],"speech":[25,50,117],"generation":[26],"tasks":[27],"remains":[28],"underexplored.":[29],"In":[30],"this":[31],"work,":[32],"we":[33,56,98],"introduce":[34],"a":[35,41,60,72,111],"text-to-speech":[36],"(TTS)":[37],"system":[38,114],"powered":[39],"by":[40],"fine-tuned":[42],"Llama":[43],"model,":[44],"named":[45],"TTS-Llama,":[46,55],"that":[47],"achieves":[48],"state-of-the-art":[49],"synthesis":[51],"performance.":[52],"Building":[53],"on":[54,82],"further":[57,99],"propose":[58],"MoLE-Llama,":[59],"text-and-speech":[61],"multimodal":[62,112],"LLM":[63],"developed":[64],"through":[65],"purely":[66],"late-fusion":[67],"parameter-efficient":[68],"fine-tuning":[69],"(PEFT)":[70],"and":[71,87],"mixture-of-expert":[73],"architecture.":[74],"Extensive":[75],"empirical":[76],"results":[77],"demonstrate":[78],"MoLE-Llama\u2019s":[79],"competitive":[80],"both":[83],"text-only":[84],"question-answering":[85],"(QA)":[86],"TTS":[88],"tasks,":[89,105],"mitigating":[90],"catastrophic":[91],"forgetting":[92],"issue":[93],"in":[94,102],"either":[95],"modality.":[96],"Finally,":[97],"explore":[100],"MoLE-Llama":[101],"text-in-speech-out":[103],"QA":[104],"demonstrating":[106],"its":[107],"great":[108],"potential":[109],"as":[110],"dialog":[113],"capable":[115],"generation.":[118]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-10-10T00:00:00"}