{"id":"https://openalex.org/W4391021666","doi":"https://doi.org/10.1109/asru57964.2023.10389705","title":"On Decoder-Only Architecture For Speech-to-Text and Large Language Model Integration","display_name":"On Decoder-Only Architecture For Speech-to-Text and Large Language Model Integration","publication_year":2023,"publication_date":"2023-12-16","ids":{"openalex":"https://openalex.org/W4391021666","doi":"https://doi.org/10.1109/asru57964.2023.10389705"},"language":"en","primary_location":{"id":"doi:10.1109/asru57964.2023.10389705","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru57964.2023.10389705","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 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/A5101674460","display_name":"Jian Wu","orcid":"https://orcid.org/0000-0002-3101-7011"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]},{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB","US"],"is_corresponding":true,"raw_author_name":"Jian Wu","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034136587","display_name":"Yashesh Gaur","orcid":null},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]},{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Yashesh Gaur","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100345092","display_name":"Zhuo Chen","orcid":"https://orcid.org/0000-0002-9011-7928"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]},{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Zhuo Chen","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106557565","display_name":"Long Zhou","orcid":"https://orcid.org/0009-0006-1919-4943"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Long Zhou","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112958005","display_name":"Yimeng Zhu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Yimeng Zhu","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106558173","display_name":"Tianrui Wang","orcid":"https://orcid.org/0000-0002-2765-5889"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]},{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Tianrui Wang","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100365053","display_name":"Jinyu Li","orcid":"https://orcid.org/0000-0002-1089-9748"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]},{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Jinyu Li","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101635405","display_name":"Shujie Liu","orcid":"https://orcid.org/0009-0008-0785-8882"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]},{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Shujie Liu","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114246785","display_name":"Bo Ren","orcid":"https://orcid.org/0000-0002-0580-6631"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Bo Ren","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030700843","display_name":"Linquan Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Linquan Liu","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101709477","display_name":"Yu Wu","orcid":"https://orcid.org/0000-0002-5715-3011"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Yu Wu","raw_affiliation_strings":["Microsoft, One Microsoft Way,Redmond,USA","Microsoft, One Microsoft Way, Redmond, USA"],"affiliations":[{"raw_affiliation_string":"Microsoft, One Microsoft Way,Redmond,USA","institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]},{"raw_affiliation_string":"Microsoft, One Microsoft Way, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":11,"corresponding_author_ids":["https://openalex.org/A5101674460"],"corresponding_institution_ids":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"],"apc_list":null,"apc_paid":null,"fwci":11.7661,"has_fulltext":false,"cited_by_count":68,"citation_normalized_percentile":{"value":0.98986183,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":100},"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":0.9983000159263611,"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":0.9983000159263611,"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":0.995199978351593,"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":0.9914000034332275,"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":0.8679794073104858},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6469094753265381},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5839400887489319},{"id":"https://openalex.org/keywords/connectionism","display_name":"Connectionism","score":0.583484947681427},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5628936886787415},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5067058205604553},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.4633849561214447},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.451190710067749},{"id":"https://openalex.org/keywords/speech-corpus","display_name":"Speech corpus","score":0.4504292607307434},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.4412499666213989},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.41652777791023254},{"id":"https://openalex.org/keywords/speech-synthesis","display_name":"Speech synthesis","score":0.3950892686843872},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.1955607831478119}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8679794073104858},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6469094753265381},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5839400887489319},{"id":"https://openalex.org/C8521452","wikidata":"https://www.wikidata.org/wiki/Q203790","display_name":"Connectionism","level":3,"score":0.583484947681427},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5628936886787415},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5067058205604553},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.4633849561214447},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.451190710067749},{"id":"https://openalex.org/C91863865","wikidata":"https://www.wikidata.org/wiki/Q4349497","display_name":"Speech corpus","level":3,"score":0.4504292607307434},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4412499666213989},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.41652777791023254},{"id":"https://openalex.org/C14999030","wikidata":"https://www.wikidata.org/wiki/Q16346","display_name":"Speech synthesis","level":2,"score":0.3950892686843872},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.1955607831478119},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asru57964.2023.10389705","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru57964.2023.10389705","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.7799999713897705,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":62,"referenced_works":["https://openalex.org/W2605131327","https://openalex.org/W2739883972","https://openalex.org/W2901607128","https://openalex.org/W2908510526","https://openalex.org/W2963250244","https://openalex.org/W2963532001","https://openalex.org/W3034625919","https://openalex.org/W3043665049","https://openalex.org/W3153583341","https://openalex.org/W3168867926","https://openalex.org/W3211278025","https://openalex.org/W4224137820","https://openalex.org/W4224308101","https://openalex.org/W4225323055","https://openalex.org/W4288089799","https://openalex.org/W4292779060","https://openalex.org/W4301581299","https://openalex.org/W4313679638","https://openalex.org/W4322718191","https://openalex.org/W4323651091","https://openalex.org/W4361866031","https://openalex.org/W4364382977","https://openalex.org/W4366330503","https://openalex.org/W4366850747","https://openalex.org/W4367628410","https://openalex.org/W4375958083","https://openalex.org/W4377372369","https://openalex.org/W4378501656","https://openalex.org/W4381827575","https://openalex.org/W4385245566","https://openalex.org/W4385823495","https://openalex.org/W4388979610","https://openalex.org/W4389524500","https://openalex.org/W4391021457","https://openalex.org/W4391021781","https://openalex.org/W4393178509","https://openalex.org/W6732953234","https://openalex.org/W6739901393","https://openalex.org/W6757817989","https://openalex.org/W6769627184","https://openalex.org/W6773820404","https://openalex.org/W6778883912","https://openalex.org/W6780680273","https://openalex.org/W6796581206","https://openalex.org/W6810081322","https://openalex.org/W6810334672","https://openalex.org/W6846175339","https://openalex.org/W6847363464","https://openalex.org/W6848735303","https://openalex.org/W6850334629","https://openalex.org/W6850625674","https://openalex.org/W6851317108","https://openalex.org/W6851513886","https://openalex.org/W6851592950","https://openalex.org/W6851847159","https://openalex.org/W6851950068","https://openalex.org/W6852326057","https://openalex.org/W6852489829","https://openalex.org/W6852818750","https://openalex.org/W6853611000","https://openalex.org/W6853998256","https://openalex.org/W6854218657"],"related_works":["https://openalex.org/W2794347674","https://openalex.org/W4200068392","https://openalex.org/W2772686614","https://openalex.org/W2152945827","https://openalex.org/W2036933852","https://openalex.org/W2015513221","https://openalex.org/W2181773877","https://openalex.org/W301864623","https://openalex.org/W2757826346","https://openalex.org/W2184371793"],"abstract_inverted_index":{"Large":[0],"language":[1,13,66],"models":[2,140],"(LLMs)":[3],"have":[4],"achieved":[5],"remarkable":[6],"success":[7],"in":[8],"the":[9,23,81,86,91,98,135],"field":[10],"of":[11,26,90,138],"natural":[12,20],"processing,":[14],"enabling":[15],"better":[16],"human-computer":[17],"interaction":[18],"using":[19],"language.":[21],"However,":[22],"seamless":[24],"integration":[25],"speech":[27,46],"signals":[28],"into":[29,63],"LLMs":[30],"has":[31,39],"not":[32,41],"been":[33,42],"explored":[34],"well.":[35],"The":[36],"\u201cdecoder-only\u201c":[37],"architecture":[38,100],"also":[40],"well":[43],"studied":[44],"for":[45,101,141],"processing":[47],"tasks.":[48],"In":[49,93],"this":[50],"research,":[51],"we":[52,95],"introduce":[53],"Speech-LLaMA,":[54],"a":[55,75,106,128],"novel":[56],"approach":[57],"that":[58],"effectively":[59],"incorporates":[60],"acoustic":[61,83],"information":[62],"text-based":[64],"large":[65],"models.":[67],"Our":[68],"method":[69],"leverages":[70],"Connectionist":[71],"Temporal":[72],"Classification":[73],"and":[74,126],"simple":[76],"audio":[77],"encoder":[78],"to":[79,85],"map":[80],"compressed":[82],"features":[84],"continuous":[87],"semantic":[88],"space":[89],"LLM.":[92],"addition,":[94],"further":[96],"probe":[97],"decoder-only":[99,139],"speech-to-text":[102,123,142],"tasks":[103,125],"by":[104],"training":[105],"smaller":[107],"scale":[108],"randomly":[109],"initialized":[110],"speech-LLaMA":[111],"model":[112],"from":[113],"speech-text":[114],"paired":[115],"data":[116],"alone.":[117],"We":[118],"conduct":[119],"experiments":[120],"on":[121],"multilingual":[122],"translation":[124],"demonstrate":[127],"significant":[129],"improvement":[130],"over":[131],"strong":[132],"baselines,":[133],"highlighting":[134],"potential":[136],"advantages":[137],"conversion.":[143]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":41},{"year":2024,"cited_by_count":21},{"year":2023,"cited_by_count":1}],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}