{"id":"https://openalex.org/W4415433014","doi":"https://doi.org/10.21437/interspeech.2025-2245","title":"How to Connect Speech Foundation Models and Large Language Models? What Matters and What Does Not","display_name":"How to Connect Speech Foundation Models and Large Language Models? What Matters and What Does Not","publication_year":2025,"publication_date":"2025-08-17","ids":{"openalex":"https://openalex.org/W4415433014","doi":"https://doi.org/10.21437/interspeech.2025-2245"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2025-2245","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2025-2245","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://hdl.handle.net/11573/1754552","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114414247","display_name":"Francesco Verdini","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Francesco Verdini","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114414249","display_name":"Pierfrancesco Melucci","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pierfrancesco Melucci","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114414250","display_name":"Stefano Perna","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Stefano Perna","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114414251","display_name":"Francesco Cariaggi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Francesco Cariaggi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054564363","display_name":"Marco Gaido","orcid":"https://orcid.org/0000-0003-4217-1396"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Marco Gaido","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070385491","display_name":"Sara Papi","orcid":"https://orcid.org/0000-0002-4494-8886"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sara Papi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114414252","display_name":"Szymon Mazurek","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Szymon Mazurek","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088995997","display_name":"Marek Kasztelnik","orcid":"https://orcid.org/0000-0002-8473-5236"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Marek Kasztelnik","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066183817","display_name":"Luisa Bentivogli","orcid":"https://orcid.org/0000-0001-7480-2231"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luisa Bentivogli","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114414248","display_name":"S\u00e9bastien Brati\u00e8res","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sebastien Brati\u00e8res","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022736687","display_name":"Paolo Merialdo","orcid":"https://orcid.org/0000-0002-3852-8092"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Paolo Merialdo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5022153057","display_name":"Simone Scardapane","orcid":"https://orcid.org/0000-0003-0881-8344"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Simone Scardapane","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1813","last_page":"1817"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.8747000098228455,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.8747000098228455,"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":0.753600001335144,"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/adapter","display_name":"Adapter (computing)","score":0.9038000106811523},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.6787999868392944},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.5976999998092651},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.49050000309944153},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.3422999978065491},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.33469998836517334}],"concepts":[{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.9038000106811523},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.6787999868392944},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6036999821662903},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5976999998092651},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.49050000309944153},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3993000090122223},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.3422999978065491},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33739998936653137},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.33469998836517334},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3140999972820282},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.31139999628067017},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.2953999936580658},{"id":"https://openalex.org/C75553542","wikidata":"https://www.wikidata.org/wiki/Q178161","display_name":"A priori and a posteriori","level":2,"score":0.27140000462532043},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.25060001015663147}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.21437/interspeech.2025-2245","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2025-2245","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2025","raw_type":"proceedings-article"},{"id":"pmh:oai:iris.uniroma1.it:11573/1754552","is_oa":true,"landing_page_url":"https://hdl.handle.net/11573/1754552","pdf_url":null,"source":{"id":"https://openalex.org/S4377196107","display_name":"IRIS Research product catalog (Sapienza University of Rome)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"pmh:oai:iris.uniroma3.it:11590/524597","is_oa":false,"landing_page_url":"https://hdl.handle.net/11590/524597","pdf_url":null,"source":{"id":"https://openalex.org/S4377196120","display_name":"Iris (Roma Tre University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I119003972","host_organization_name":"Roma Tre University","host_organization_lineage":["https://openalex.org/I119003972"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"pmh:oai:iris.uniroma1.it:11573/1754552","is_oa":true,"landing_page_url":"https://hdl.handle.net/11573/1754552","pdf_url":null,"source":{"id":"https://openalex.org/S4377196107","display_name":"IRIS Research product catalog (Sapienza University of Rome)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"remarkable":[1],"performance":[2,65],"achieved":[3],"by":[4],"Large":[5],"Language":[6],"Models":[7],"(LLM)":[8],"has":[9,58,140],"driven":[10],"research":[11],"efforts":[12],"to":[13],"leverage":[14],"them":[15],"for":[16],"a":[17,41,130],"wide":[18],"range":[19],"of":[20,33,37,40,78,96],"tasks":[21],"and":[22,86,103,105,109,120,143,148],"input":[23],"modalities.":[24],"In":[25],"speech-to-text":[26],"(S2T)":[27],"tasks,":[28,115],"the":[29,35,38,47,63,75,79,83,94,127,137,146],"emerging":[30],"solution":[31],"consists":[32],"projecting":[34],"output":[36],"encoder":[39],"Speech":[42,118,121],"Foundational":[43],"Model":[44],"(SFM)":[45],"into":[46],"LLM":[48],"embedding":[49],"space":[50],"through":[51],"an":[52],"adapter":[53,80,98,138],"module.":[54],"However,":[55],"no":[56],"work":[57],"yet":[59],"investigated":[60],"how":[61],"much":[62],"downstream-task":[64],"depends":[66,81,144],"on":[67,82,111,145],"each":[68],"component":[69],"(SFM,":[70],"adapter,":[71],"LLM)":[72],"nor":[73],"whether":[74],"best":[76],"design":[77],"chosen":[84],"SFM":[85,128,147],"LLM.":[87,149],"To":[88],"fill":[89],"this":[90],"gap,":[91],"we":[92],"evaluate":[93],"combination":[95],"5":[97],"modules,":[99],"2":[100,106],"LLMs":[101],"(Mistral":[102],"Llama),":[104],"SFMs":[107],"(Whisper":[108],"SeamlessM4T)":[110],"two":[112],"widespread":[113],"S2T":[114],"namely":[116],"Automatic":[117],"Recognition":[119],"Translation.":[122],"Our":[123],"results":[124],"demonstrate":[125],"that":[126],"plays":[129],"pivotal":[131],"role":[132],"in":[133],"downstream":[134],"performance,":[135],"while":[136],"choice":[139],"moderate":[141],"impact":[142]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-24T00:00:00"}