{"id":"https://openalex.org/W4406461487","doi":"https://doi.org/10.1109/slt61566.2024.10832146","title":"Bestow: Efficient and Streamable Speech Language Model with The Best of Two Worlds in GPT and T5","display_name":"Bestow: Efficient and Streamable Speech Language Model with The Best of Two Worlds in GPT and T5","publication_year":2024,"publication_date":"2024-12-02","ids":{"openalex":"https://openalex.org/W4406461487","doi":"https://doi.org/10.1109/slt61566.2024.10832146"},"language":"en","primary_location":{"id":"doi:10.1109/slt61566.2024.10832146","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt61566.2024.10832146","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE Spoken Language Technology Workshop (SLT)","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/A5002433660","display_name":"Zhehuai Chen","orcid":"https://orcid.org/0000-0003-4400-5340"},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhehuai Chen","raw_affiliation_strings":["NVIDIA,Santa Clara,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA,Santa Clara,CA,USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100411301","display_name":"He Huang","orcid":"https://orcid.org/0000-0002-9217-4977"},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"He Huang","raw_affiliation_strings":["NVIDIA,Santa Clara,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA,Santa Clara,CA,USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088392468","display_name":"Oleksii Hrinchuk","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Oleksii Hrinchuk","raw_affiliation_strings":["NVIDIA,Santa Clara,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA,Santa Clara,CA,USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069691915","display_name":"Krishna C. Puvvada","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Krishna C. Puvvada","raw_affiliation_strings":["NVIDIA,Santa Clara,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA,Santa Clara,CA,USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040007143","display_name":"Nithin Rao Koluguri","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nithin Rao Koluguri","raw_affiliation_strings":["NVIDIA,Santa Clara,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA,Santa Clara,CA,USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027217976","display_name":"Piotr \u017belasko","orcid":"https://orcid.org/0000-0002-8245-0413"},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Piotr \u017belasko","raw_affiliation_strings":["NVIDIA,Santa Clara,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA,Santa Clara,CA,USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040747392","display_name":"Jagadeesh Balam","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jagadeesh Balam","raw_affiliation_strings":["NVIDIA,Santa Clara,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA,Santa Clara,CA,USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5032957280","display_name":"Boris Ginsburg","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Boris Ginsburg","raw_affiliation_strings":["NVIDIA,Santa Clara,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA,Santa Clara,CA,USA","institution_ids":["https://openalex.org/I4210127875"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210127875"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"147","last_page":"154"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9993000030517578,"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":0.9993000030517578,"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":0.9976000189781189,"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":0.9973000288009644,"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":0.6539334058761597},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4152681827545166},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.3604622483253479},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.3600345253944397},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.33320990204811096},{"id":"https://openalex.org/keywords/philosophy","display_name":"Philosophy","score":0.07997015118598938}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6539334058761597},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4152681827545166},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3604622483253479},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3600345253944397},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33320990204811096},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.07997015118598938}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/slt61566.2024.10832146","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt61566.2024.10832146","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.5199999809265137,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":76,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2327501763","https://openalex.org/W2951534261","https://openalex.org/W2974231335","https://openalex.org/W2995181338","https://openalex.org/W3016010032","https://openalex.org/W3030437843","https://openalex.org/W3097777922","https://openalex.org/W3168867926","https://openalex.org/W3196509775","https://openalex.org/W3198429080","https://openalex.org/W3198694222","https://openalex.org/W4225323055","https://openalex.org/W4287855189","https://openalex.org/W4288089799","https://openalex.org/W4292779060","https://openalex.org/W4296068815","https://openalex.org/W4297727296","https://openalex.org/W4311000453","https://openalex.org/W4319862635","https://openalex.org/W4323066695","https://openalex.org/W4366330503","https://openalex.org/W4381827575","https://openalex.org/W4385245566","https://openalex.org/W4385572700","https://openalex.org/W4385822439","https://openalex.org/W4386755195","https://openalex.org/W4387891768","https://openalex.org/W4388891078","https://openalex.org/W4389500133","https://openalex.org/W4389520395","https://openalex.org/W4389524500","https://openalex.org/W4389600306","https://openalex.org/W4390041933","https://openalex.org/W4390632065","https://openalex.org/W4391021542","https://openalex.org/W4391021623","https://openalex.org/W4391021627","https://openalex.org/W4391021666","https://openalex.org/W4391591726","https://openalex.org/W4392902623","https://openalex.org/W4392903872","https://openalex.org/W4392903956","https://openalex.org/W4392931626","https://openalex.org/W4399758690","https://openalex.org/W4402111234","https://openalex.org/W4402111955","https://openalex.org/W4402112192","https://openalex.org/W4402671881","https://openalex.org/W4404781802","https://openalex.org/W6764357534","https://openalex.org/W6767671539","https://openalex.org/W6769627184","https://openalex.org/W6771467084","https://openalex.org/W6771812881","https://openalex.org/W6778883912","https://openalex.org/W6796581206","https://openalex.org/W6810334672","https://openalex.org/W6810759146","https://openalex.org/W6838912845","https://openalex.org/W6847363464","https://openalex.org/W6850218400","https://openalex.org/W6851592950","https://openalex.org/W6853998256","https://openalex.org/W6855414158","https://openalex.org/W6856561546","https://openalex.org/W6856689524","https://openalex.org/W6856800273","https://openalex.org/W6857054612","https://openalex.org/W6858930701","https://openalex.org/W6859467570","https://openalex.org/W6859810418","https://openalex.org/W6860041859","https://openalex.org/W6860549370","https://openalex.org/W6861859833","https://openalex.org/W6869902183"],"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/W3204019825"],"abstract_inverted_index":{"Incorporating":[0],"speech":[1,25,44,97,151],"understanding":[2],"capabilities":[3],"into":[4,67],"pretrained":[5,51],"large-language":[6],"models":[7],"has":[8,76],"become":[9],"a":[10,32,38,68,104,147],"vital":[11],"research":[12,114],"direction":[13],"(SpeechLLM).":[14],"The":[15],"previous":[16],"architectures":[17],"can":[18],"be":[19],"categorized":[20],"as:":[21],"i)":[22],"GPT-style,":[23],"prepend":[24],"prompts":[26,30],"to":[27,46,57,96,171],"the":[28,50,59,92,110,121,136],"text":[29],"as":[31,103],"sequence":[33],"of":[34,49,150],"LLM":[35,168],"inputs":[36],"like":[37],"decoder-only":[39],"model;":[40],"ii)":[41],"T5-style,":[42],"introduce":[43],"cross-attention":[45],"each":[47],"layer":[48],"LLMs.":[52],"We":[53,99],"propose":[54],"BESTOW":[55,116],"architecture":[56],"bring":[58],"BESt":[60],"features":[61],"from":[62],"$T":[63],"w":[64],"O$":[65],"Worlds":[66],"single":[69],"model":[70],"that":[71,126],"is":[72,82,159],"highly":[73],"efficient":[74],"and":[75,108,112,129,166],"strong":[77,144],"multitask":[78],"capabilities.":[79],"Moreover,":[80],"there":[81],"no":[83],"clear":[84],"streaming":[85,113],"solution":[86,93,125,141],"for":[87],"either":[88],"style,":[89],"especially":[90],"considering":[91],"should":[94],"generalize":[95],"multitask.":[98],"reformulate":[100],"streamable":[101,140],"SpeechLLM":[102,124],"read-write":[105],"policy":[106],"problem":[107],"unifies":[109],"offline":[111],"with":[115,162],"architecture.":[117],"Hence":[118],"we":[119],"demonstrate":[120],"first":[122],"open-source":[123],"enables":[127],"Streaming":[128],"Multitask":[130],"at":[131,135],"scale":[132],"(beyond":[133],"ASR)":[134],"same":[137],"time.":[138],"This":[139],"achieves":[142],"very":[143],"performance":[145],"on":[146],"wide":[148],"range":[149],"tasks":[152],"(ASR,":[153],"AST,":[154],"SQA,":[155],"unseen":[156],"DynamicSuperb).":[157],"It":[158],"end-to-end":[160],"optimizable,":[161],"lower":[163],"training/inference":[164],"cost,":[165],"demonstrates":[167],"knowledge":[169],"transferability":[170],"speech.":[172]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":4}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}