{"id":"https://openalex.org/W4391021623","doi":"https://doi.org/10.1109/asru57964.2023.10389703","title":"SLM: Bridge the Thin Gap Between Speech and Text Foundation Models","display_name":"SLM: Bridge the Thin Gap Between Speech and Text Foundation Models","publication_year":2023,"publication_date":"2023-12-16","ids":{"openalex":"https://openalex.org/W4391021623","doi":"https://doi.org/10.1109/asru57964.2023.10389703"},"language":"en","primary_location":{"id":"doi:10.1109/asru57964.2023.10389703","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru57964.2023.10389703","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/A5036732709","display_name":"Mingqiu Wang","orcid":"https://orcid.org/0000-0001-7164-0054"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Mingqiu Wang","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100750907","display_name":"Wei Han","orcid":"https://orcid.org/0000-0002-4201-9645"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wei Han","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025419994","display_name":"Izhak Shafran","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Izhak Shafran","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035558261","display_name":"Zelin Wu","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zelin Wu","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027763497","display_name":"Chung\u2010Cheng Chiu","orcid":"https://orcid.org/0000-0001-9729-4778"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chung-Cheng Chiu","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101989491","display_name":"Yuan Cao","orcid":"https://orcid.org/0000-0002-2495-4701"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yuan Cao","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071702566","display_name":"Nanxin Chen","orcid":"https://orcid.org/0000-0001-6698-1604"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nanxin Chen","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100433630","display_name":"Yu Zhang","orcid":"https://orcid.org/0000-0002-8345-3835"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yu Zhang","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010878109","display_name":"Hagen Soltau","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hagen Soltau","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026258361","display_name":"Paul K. Rubenstein","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Paul K. Rubenstein","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031888392","display_name":"Luk\u00e1\u0161 \u017dilka","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lukas Zilka","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101834699","display_name":"Dian Yu","orcid":"https://orcid.org/0000-0002-8583-8931"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dian Yu","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051090720","display_name":"Golan Pundak","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Golan Pundak","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057946983","display_name":"Nikhil Siddhartha","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nikhil Siddhartha","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081836495","display_name":"Johan Schalkwyk","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Johan Schalkwyk","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5010253402","display_name":"Yonghui Wu","orcid":"https://orcid.org/0000-0002-6780-6135"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yonghui Wu","raw_affiliation_strings":["Google Deepmind"],"affiliations":[{"raw_affiliation_string":"Google Deepmind","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":16,"corresponding_author_ids":["https://openalex.org/A5036732709"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":4.7748,"has_fulltext":false,"cited_by_count":28,"citation_normalized_percentile":{"value":0.96098035,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"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/T10028","display_name":"Topic Modeling","score":0.9998999834060669,"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/T10028","display_name":"Topic Modeling","score":0.9998999834060669,"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":0.9998999834060669,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9991999864578247,"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":0.8559137582778931},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6267260313034058},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5127950310707092},{"id":"https://openalex.org/keywords/bridge","display_name":"Bridge (graph theory)","score":0.4724752604961395},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4497615098953247},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4380796253681183},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.43200698494911194},{"id":"https://openalex.org/keywords/dialog-box","display_name":"Dialog box","score":0.418517529964447}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8559137582778931},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6267260313034058},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5127950310707092},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.4724752604961395},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4497615098953247},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4380796253681183},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.43200698494911194},{"id":"https://openalex.org/C173853756","wikidata":"https://www.wikidata.org/wiki/Q86915","display_name":"Dialog box","level":2,"score":0.418517529964447},{"id":"https://openalex.org/C126322002","wikidata":"https://www.wikidata.org/wiki/Q11180","display_name":"Internal medicine","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.0},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asru57964.2023.10389703","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru57964.2023.10389703","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":[{"score":0.49000000953674316,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":50,"referenced_works":["https://openalex.org/W38194800","https://openalex.org/W100623710","https://openalex.org/W2101105183","https://openalex.org/W2896457183","https://openalex.org/W2912924812","https://openalex.org/W2963532001","https://openalex.org/W3091928890","https://openalex.org/W3119308075","https://openalex.org/W3139918052","https://openalex.org/W3168867926","https://openalex.org/W3169483174","https://openalex.org/W3196509775","https://openalex.org/W3197324626","https://openalex.org/W3209059054","https://openalex.org/W4221155340","https://openalex.org/W4226120743","https://openalex.org/W4288089799","https://openalex.org/W4307079201","https://openalex.org/W4319862232","https://openalex.org/W4319862635","https://openalex.org/W4323066695","https://openalex.org/W4377130946","https://openalex.org/W4377372369","https://openalex.org/W4381786045","https://openalex.org/W4381827575","https://openalex.org/W4385571124","https://openalex.org/W4385573964","https://openalex.org/W4385822589","https://openalex.org/W4385822641","https://openalex.org/W4386071707","https://openalex.org/W4389524500","https://openalex.org/W6755207826","https://openalex.org/W6759579507","https://openalex.org/W6769627184","https://openalex.org/W6784545093","https://openalex.org/W6784577980","https://openalex.org/W6791904447","https://openalex.org/W6796581206","https://openalex.org/W6810259195","https://openalex.org/W6810334672","https://openalex.org/W6810673746","https://openalex.org/W6845943790","https://openalex.org/W6847076894","https://openalex.org/W6847363464","https://openalex.org/W6847568899","https://openalex.org/W6850218400","https://openalex.org/W6852800892","https://openalex.org/W6852818750","https://openalex.org/W6853249747","https://openalex.org/W6853998256"],"related_works":["https://openalex.org/W2098987383","https://openalex.org/W2417260800","https://openalex.org/W2795961259","https://openalex.org/W1596203174","https://openalex.org/W2117933979","https://openalex.org/W4298396513","https://openalex.org/W2283130723","https://openalex.org/W103938586","https://openalex.org/W2104718772","https://openalex.org/W4233992201"],"abstract_inverted_index":{"We":[0],"present":[1,162],"a":[2,9,39,89,93,142,147],"joint":[3],"Speech":[4],"and":[5,12,22,36,70,92,115,128,137],"Language":[6],"Model":[7],"(SLM),":[8],"multitask,":[10],"multilingual,":[11],"dual-modal":[13],"model":[14],"that":[15,121],"takes":[16],"advantage":[17],"of":[18,46,81,166],"pretrained":[19,28,126],"foundational":[20],"speech":[21,67,72,90,113,127],"language":[23,129],"models.":[24],"SLM":[25,56,96,149],"freezes":[26],"the":[27,47,78,122],"foundation":[29,48,164],"models":[30,130,165],"to":[31,57,99,154],"maximally":[32],"preserves":[33],"their":[34],"capabilities,":[35],"only":[37,54,152],"trains":[38],"simple":[40,143],"adapter":[41],"with":[42],"just":[43],"1%":[44],"(156M)":[45],"models\u2019":[49],"parameters.":[50],"This":[51],"adaptation":[52,144],"not":[53,151],"leads":[55],"achieve":[58],"strong":[59,159],"performance":[60],"on":[61],"conventional":[62],"tasks":[63,103],"such":[64],"as":[65],"automatic":[66,71],"recognition":[68],"(ASR)":[69],"translation":[73],"(AST),":[74],"but":[75,156],"also":[76,157],"unlocks":[77],"novel":[79],"capability":[80],"zero-shot":[82],"instruction-following":[83],"for":[84],"more":[85],"diverse":[86],"tasks.":[87],"Given":[88],"input":[91],"text":[94],"instruction,":[95],"is":[97,131,150],"able":[98],"perform":[100],"unseen":[101],"generation":[102],"including":[104],"contextual":[105],"biasing":[106],"ASR":[107],"using":[108],"real-time":[109],"context,":[110],"dialog":[111],"generation,":[112],"continuation,":[114],"question":[116],"answering.":[117],"Our":[118],"approach":[119],"demonstrates":[120],"representational":[123],"gap":[124],"between":[125],"narrower":[132],"than":[133],"one":[134],"would":[135],"expect,":[136],"can":[138],"be":[139],"bridged":[140],"by":[141],"mechanism.":[145],"As":[146],"result,":[148],"efficient":[153],"train,":[155],"inherits":[158],"capabilities":[160],"already":[161],"in":[163],"different":[167],"modalities.":[168]},"counts_by_year":[{"year":2026,"cited_by_count":6},{"year":2025,"cited_by_count":16},{"year":2024,"cited_by_count":6}],"updated_date":"2026-04-23T09:07:50.710637","created_date":"2025-10-10T00:00:00"}