{"id":"https://openalex.org/W4392909554","doi":"https://doi.org/10.1109/icassp48485.2024.10448115","title":"Training Audio Captioning Models without Audio","display_name":"Training Audio Captioning Models without Audio","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392909554","doi":"https://doi.org/10.1109/icassp48485.2024.10448115"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10448115","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10448115","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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":"https://openalex.org/A5017946811","display_name":"Soham Deshmukh","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"]}],"countries":["GB"],"is_corresponding":true,"raw_author_name":"Soham Deshmukh","raw_affiliation_strings":["Microsoft"],"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073250019","display_name":"Benjamin Elizalde","orcid":"https://orcid.org/0000-0001-6461-5790"},"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"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Benjamin Elizalde","raw_affiliation_strings":["Microsoft"],"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034420214","display_name":"Dimitra Emmanouilidou","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"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Dimitra Emmanouilidou","raw_affiliation_strings":["Microsoft"],"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113017615","display_name":"Bhiksha Raj","orcid":null},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bhiksha Raj","raw_affiliation_strings":["Carnegie Mellon University"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102775511","display_name":"Rita Singh","orcid":"https://orcid.org/0000-0003-3743-0162"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rita Singh","raw_affiliation_strings":["Carnegie Mellon University"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101528071","display_name":"Huaming Wang","orcid":"https://orcid.org/0000-0003-1490-2673"},"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"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Huaming Wang","raw_affiliation_strings":["Microsoft"],"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5017946811"],"corresponding_institution_ids":["https://openalex.org/I4210164937"],"apc_list":null,"apc_paid":null,"fwci":6.0287,"has_fulltext":false,"cited_by_count":16,"citation_normalized_percentile":{"value":0.97044493,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"371","last_page":"375"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T11309","display_name":"Music and Audio Processing","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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":0.9983999729156494,"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":0.9955000281333923,"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/closed-captioning","display_name":"Closed captioning","score":0.897567868232727},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8439910411834717},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6232868432998657},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5052321553230286},{"id":"https://openalex.org/keywords/audio-mining","display_name":"Audio mining","score":0.47695791721343994},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.44830384850502014},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.44442301988601685},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4271549582481384},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.39693716168403625},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.262481153011322},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.1744656264781952}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.897567868232727},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8439910411834717},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6232868432998657},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5052321553230286},{"id":"https://openalex.org/C157968479","wikidata":"https://www.wikidata.org/wiki/Q3079876","display_name":"Audio mining","level":4,"score":0.47695791721343994},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.44830384850502014},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.44442301988601685},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4271549582481384},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39693716168403625},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.262481153011322},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.1744656264781952},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"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/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10448115","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10448115","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W2130942839","https://openalex.org/W2995348821","https://openalex.org/W3015591594","https://openalex.org/W3034999214","https://openalex.org/W3094550259","https://openalex.org/W3112467147","https://openalex.org/W3162331882","https://openalex.org/W3204936728","https://openalex.org/W3205708381","https://openalex.org/W3206996142","https://openalex.org/W4225432580","https://openalex.org/W4226442948","https://openalex.org/W4309129408","https://openalex.org/W4318718996","https://openalex.org/W4318752004","https://openalex.org/W4321473516","https://openalex.org/W4372260310","https://openalex.org/W4372266552","https://openalex.org/W4372340819","https://openalex.org/W4377372369","https://openalex.org/W4385567053","https://openalex.org/W4385822467","https://openalex.org/W4392903801","https://openalex.org/W4400033239","https://openalex.org/W6679436768","https://openalex.org/W6787564935","https://openalex.org/W6799303324","https://openalex.org/W6801865844","https://openalex.org/W6802577842","https://openalex.org/W6810452849","https://openalex.org/W6846793904","https://openalex.org/W6849109464","https://openalex.org/W6849416043","https://openalex.org/W6850311597","https://openalex.org/W6852818750"],"related_works":["https://openalex.org/W2539919382","https://openalex.org/W22517275","https://openalex.org/W2042419400","https://openalex.org/W2122924390","https://openalex.org/W2064347532","https://openalex.org/W2131711534","https://openalex.org/W2589081144","https://openalex.org/W2171620991","https://openalex.org/W189256204","https://openalex.org/W2293401412"],"abstract_inverted_index":{"Automated":[0],"Audio":[1],"Captioning":[2],"(AAC)":[3],"is":[4,39,100,152],"the":[5,47,71,90,97,103,110,120,134],"task":[6],"of":[7,25,35,74,122],"generating":[8],"natural":[9],"language":[10],"descriptions":[11],"given":[12],"an":[13,59],"audio":[14,26,106,116,159,167],"stream.":[15],"A":[16],"typical":[17],"AAC":[18,63],"system":[19],"requires":[20],"manually":[21],"curated":[22],"training":[23,165],"data":[24,44],"segments":[27],"and":[28,57,115,161],"corresponding":[29],"text":[30,93,98,114,170],"caption":[31,162],"annotations.":[32],"The":[33],"creation":[34],"these":[36],"audio-caption":[37],"pairs":[38],"costly,":[40],"resulting":[41],"in":[42],"general":[43],"scarcity":[45],"for":[46],"task.":[48],"In":[49],"this":[50,54],"work,":[51],"we":[52,118,155],"address":[53],"major":[55],"limitation":[56],"propose":[58,119],"approach":[60,69],"to":[61],"train":[62],"systems":[64],"using":[65],"only":[66],"text.":[67],"Our":[68],"leverages":[70],"multimodal":[72],"space":[73],"contrastively":[75],"trained":[76,143],"audio-text":[77],"models,":[78],"such":[79],"as":[80],"CLAP.":[81],"During":[82,95],"training,":[83],"a":[84,126],"decoder":[85],"generates":[86],"captions":[87],"conditioned":[88],"on":[89],"pretrained":[91,104],"CLAP":[92,105],"encoder.":[94,107],"inference,":[96],"encoder":[99],"replaced":[101],"with":[102,140,144],"To":[108],"bridge":[109],"modality":[111],"gap":[112],"between":[113],"embeddings,":[117],"use":[121],"noise":[123],"injection":[124],"or":[125,168],"learnable":[127],"adapter,":[128],"during":[129],"training.":[130],"We":[131],"find":[132],"that":[133,148],"proposed":[135],"text-only":[136],"framework":[137],"performs":[138],"competitively":[139],"stateof-the-art":[141],"models":[142],"paired":[145],"audio,":[146],"showing":[147],"efficient":[149],"text-to-audio":[150],"transfer":[151],"possible.":[153],"Finally,":[154],"showcase":[156],"both":[157],"stylized":[158],"captioning":[160],"enrichment":[163],"while":[164],"without":[166],"human-created":[169],"captions.":[171]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":10},{"year":2024,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}