{"id":"https://openalex.org/W4392904208","doi":"https://doi.org/10.1109/icassp48485.2024.10448486","title":"A Sound Approach: Using Large Language Models to Generate Audio Descriptions for Egocentric Text-Audio Retrieval","display_name":"A Sound Approach: Using Large Language Models to Generate Audio Descriptions for Egocentric Text-Audio Retrieval","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392904208","doi":"https://doi.org/10.1109/icassp48485.2024.10448486"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10448486","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10448486","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":"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/A5036345396","display_name":"Andreea-Maria Oncescu","orcid":"https://orcid.org/0000-0001-9730-3873"},"institutions":[{"id":"https://openalex.org/I40120149","display_name":"University of Oxford","ror":"https://ror.org/052gg0110","country_code":"GB","type":"education","lineage":["https://openalex.org/I40120149"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Andreea-Maria Oncescu","raw_affiliation_strings":["University of Oxford"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Oxford","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109730096","display_name":"Jo\u00e3o F. Henriques","orcid":"https://orcid.org/0000-0002-2478-2102"},"institutions":[{"id":"https://openalex.org/I40120149","display_name":"University of Oxford","ror":"https://ror.org/052gg0110","country_code":"GB","type":"education","lineage":["https://openalex.org/I40120149"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Jo\u00e3o F. Henriques","raw_affiliation_strings":["University of Oxford"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Oxford","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057678172","display_name":"Andrew Zisserman","orcid":"https://orcid.org/0000-0002-8945-8573"},"institutions":[{"id":"https://openalex.org/I40120149","display_name":"University of Oxford","ror":"https://ror.org/052gg0110","country_code":"GB","type":"education","lineage":["https://openalex.org/I40120149"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Andrew Zisserman","raw_affiliation_strings":["University of Oxford"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Oxford","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102018867","display_name":"Samuel Albanie","orcid":"https://orcid.org/0000-0003-1732-9198"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Samuel Albanie","raw_affiliation_strings":["University of Cambridge"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Cambridge","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001366802","display_name":"A. Sophia Koepke","orcid":"https://orcid.org/0000-0002-5807-0576"},"institutions":[{"id":"https://openalex.org/I143910747","display_name":"TH Bingen University of Applied Sciences","ror":"https://ror.org/01pxkj057","country_code":"DE","type":"education","lineage":["https://openalex.org/I143910747"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"A. Sophia Koepke","raw_affiliation_strings":["University of Tübingen"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Tübingen","institution_ids":["https://openalex.org/I143910747"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.6686,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.83091392,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"7300","last_page":"7304"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9994999766349792,"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":0.9994999766349792,"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/T10515","display_name":"Cancer-related molecular mechanisms research","score":0.9878000020980835,"subfield":{"id":"https://openalex.org/subfields/1306","display_name":"Cancer Research"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9829999804496765,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.8642162680625916},{"id":"https://openalex.org/keywords/audio-visual","display_name":"Audio visual","score":0.7222537398338318},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.5729120373725891},{"id":"https://openalex.org/keywords/audio-mining","display_name":"Audio mining","score":0.5313248038291931},{"id":"https://openalex.org/keywords/the-internet","display_name":"The Internet","score":0.4859398603439331},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.4426637887954712},{"id":"https://openalex.org/keywords/audio-signal-processing","display_name":"Audio signal processing","score":0.42689406871795654},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.415072500705719},{"id":"https://openalex.org/keywords/audio-signal","display_name":"Audio signal","score":0.3857662081718445},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.38321810960769653},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3738606572151184},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.35581690073013306},{"id":"https://openalex.org/keywords/multimedia","display_name":"Multimedia","score":0.31030625104904175},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.22989517450332642},{"id":"https://openalex.org/keywords/world-wide-web","display_name":"World Wide Web","score":0.18732857704162598},{"id":"https://openalex.org/keywords/speech-coding","display_name":"Speech coding","score":0.1851070523262024},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.12429571151733398}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8642162680625916},{"id":"https://openalex.org/C3017588708","wikidata":"https://www.wikidata.org/wiki/Q758901","display_name":"Audio visual","level":2,"score":0.7222537398338318},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.5729120373725891},{"id":"https://openalex.org/C157968479","wikidata":"https://www.wikidata.org/wiki/Q3079876","display_name":"Audio mining","level":4,"score":0.5313248038291931},{"id":"https://openalex.org/C110875604","wikidata":"https://www.wikidata.org/wiki/Q75","display_name":"The Internet","level":2,"score":0.4859398603439331},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.4426637887954712},{"id":"https://openalex.org/C127220857","wikidata":"https://www.wikidata.org/wiki/Q2719318","display_name":"Audio signal processing","level":4,"score":0.42689406871795654},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.415072500705719},{"id":"https://openalex.org/C64922751","wikidata":"https://www.wikidata.org/wiki/Q4650799","display_name":"Audio signal","level":3,"score":0.3857662081718445},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.38321810960769653},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3738606572151184},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.35581690073013306},{"id":"https://openalex.org/C49774154","wikidata":"https://www.wikidata.org/wiki/Q131765","display_name":"Multimedia","level":1,"score":0.31030625104904175},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.22989517450332642},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.18732857704162598},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.1851070523262024},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.12429571151733398},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp48485.2024.10448486","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10448486","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"},{"id":"pmh:oai:ora.ox.ac.uk:uuid:ac6fc325-6ec9-422d-8034-01607559a300","is_oa":false,"landing_page_url":"https://ora.ox.ac.uk/objects/uuid:ac6fc325-6ec9-422d-8034-01607559a300","pdf_url":null,"source":{"id":"https://openalex.org/S4306402636","display_name":"Oxford University Research Archive (ORA) (University of Oxford)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I40120149","host_organization_name":"University of Oxford","host_organization_lineage":["https://openalex.org/I40120149"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Symplectic Elements","raw_type":"Conference item"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.46000000834465027,"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10"}],"awards":[{"id":"https://openalex.org/G1341008233","display_name":null,"funder_award_id":"EXC 2064/1","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G5003611146","display_name":"Visual AI: An Open World Interpretable Visual Transformer","funder_award_id":"EP/T028572/1","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G7907426250","display_name":"EXC 2064: Maschinelles Lernen: Neue Perspektiven f\u00fcr die Wissenschaft","funder_award_id":"390727645","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G8060345762","display_name":null,"funder_award_id":"RF\\201819\\18\\163","funder_id":"https://openalex.org/F4320320005","funder_display_name":"Royal Academy of Engineering"},{"id":"https://openalex.org/G8128963897","display_name":null,"funder_award_id":"EXC 2064","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G8161891803","display_name":null,"funder_award_id":"EP/T028572/1","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G8556661049","display_name":null,"funder_award_id":"390727645","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G8915364236","display_name":null,"funder_award_id":"2064/1","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"}],"funders":[{"id":"https://openalex.org/F4320320005","display_name":"Royal Academy of Engineering","ror":"https://ror.org/0526snb40"},{"id":"https://openalex.org/F4320320879","display_name":"Deutsche Forschungsgemeinschaft","ror":"https://ror.org/018mejw64"},{"id":"https://openalex.org/F4320322909","display_name":"Isaac Newton Trust","ror":"https://ror.org/02gn6ta77"},{"id":"https://openalex.org/F4320334627","display_name":"Engineering and Physical Sciences Research Council","ror":"https://ror.org/0439y7842"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W2069870183","https://openalex.org/W2796136333","https://openalex.org/W2961193895","https://openalex.org/W2971680695","https://openalex.org/W3015591594","https://openalex.org/W3146463283","https://openalex.org/W3198452188","https://openalex.org/W3205786327","https://openalex.org/W3206366488","https://openalex.org/W3207758636","https://openalex.org/W4210913346","https://openalex.org/W4226442948","https://openalex.org/W4297950672","https://openalex.org/W4320086121","https://openalex.org/W4320813149","https://openalex.org/W4322718191","https://openalex.org/W4361866031","https://openalex.org/W4366850747","https://openalex.org/W4372260217","https://openalex.org/W4372260310","https://openalex.org/W4380352233","https://openalex.org/W4384918448","https://openalex.org/W4386076314","https://openalex.org/W4386083063","https://openalex.org/W4390873954","https://openalex.org/W4400033239","https://openalex.org/W6749916090","https://openalex.org/W6843825870","https://openalex.org/W6849976536","https://openalex.org/W6849990444","https://openalex.org/W6850625674","https://openalex.org/W6851513886","https://openalex.org/W6851950068","https://openalex.org/W6853690301","https://openalex.org/W6854866820"],"related_works":["https://openalex.org/W2547431347","https://openalex.org/W2289868279","https://openalex.org/W4231351862","https://openalex.org/W4212794605","https://openalex.org/W4315836293","https://openalex.org/W1611900921","https://openalex.org/W2064012922","https://openalex.org/W1620668332","https://openalex.org/W2033602688","https://openalex.org/W2156141367"],"abstract_inverted_index":{"Video":[0],"databases":[1],"from":[2,34,60],"the":[3,24,81,94,101,117,126,136,143,149,161,165],"internet":[4],"are":[5,41,45],"a":[6,65,169],"valuable":[7],"source":[8],"of":[9,23,148,163],"text-audio":[10,53,89],"retrieval":[11,90,137],"datasets.":[12],"However,":[13],"given":[14],"that":[15,124,154],"sound":[16],"and":[17,85,96,99],"vision":[18],"streams":[19],"represent":[20],"different":[21],"\"views\"":[22],"data,":[25],"treating":[26],"visual":[27],"descriptions":[28,31,70,109],"as":[29],"audio":[30,38,58,145],"is":[32],"far":[33],"optimal.":[35],"Even":[36],"if":[37],"class":[39,146],"labels":[40,147],"present,":[42],"they":[43],"commonly":[44],"not":[46],"very":[47],"detailed,":[48],"making":[49],"them":[50],"unsuited":[51],"for":[52,67,106],"retrieval.":[54],"To":[55],"exploit":[56],"relevant":[57],"information":[59],"video-text":[61],"datasets,":[62],"we":[63,79,122,129,152],"introduce":[64],"methodology":[66],"generating":[68],"audio-centric":[69,108],"using":[71,116,125,142],"Large":[72],"Language":[73],"Models":[74],"(LLMs).":[75],"In":[76],"this":[77],"work,":[78],"consider":[80],"egocentric":[82],"video":[83],"setting":[84],"propose":[86],"three":[87],"new":[88],"benchmarks":[91],"based":[92],"on":[93,100,138],"EpicMIR":[95],"EgoMCQ":[97],"tasks,":[98],"EpicSounds":[102],"dataset.":[103,150],"Our":[104],"approach":[105],"obtaining":[107],"gives":[110],"significantly":[111],"higher":[112],"zero-shot":[113],"performance":[114],"than":[115],"original":[118,144],"visual-centric":[119],"descriptions.":[120],"Furthermore,":[121],"show":[123],"same":[127],"prompts,":[128],"can":[130,156],"successfully":[131],"employ":[132],"LLMs":[133,155],"to":[134,141,159],"improve":[135],"EpicSounds,":[139],"compared":[140],"Finally,":[151],"confirm":[153],"be":[157],"used":[158],"determine":[160],"difficulty":[162],"identifying":[164],"action":[166],"associated":[167],"with":[168],"sound.":[170]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":3}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}