{"id":"https://openalex.org/W4392903033","doi":"https://doi.org/10.1109/icassp48485.2024.10447215","title":"Improving Audio Captioning Models with Fine-Grained Audio Features, Text Embedding Supervision, and LLM Mix-Up Augmentation","display_name":"Improving Audio Captioning Models with Fine-Grained Audio Features, Text Embedding Supervision, and LLM Mix-Up Augmentation","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392903033","doi":"https://doi.org/10.1109/icassp48485.2024.10447215"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10447215","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447215","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/A5088967620","display_name":"Shih-Lun Wu","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":true,"raw_author_name":"Shih-Lun Wu","raw_affiliation_strings":["Carnegie Mellon University,Language Technologies Institute,Pittsburgh,PA,USA","Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA, USA"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University,Language Technologies Institute,Pittsburgh,PA,USA","institution_ids":["https://openalex.org/I74973139"]},{"raw_affiliation_string":"Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050058892","display_name":"Xuankai Chang","orcid":"https://orcid.org/0000-0002-5221-5412"},"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":"Xuankai Chang","raw_affiliation_strings":["Carnegie Mellon University,Language Technologies Institute,Pittsburgh,PA,USA","Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA, USA"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University,Language Technologies Institute,Pittsburgh,PA,USA","institution_ids":["https://openalex.org/I74973139"]},{"raw_affiliation_string":"Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086940921","display_name":"Gordon Wichern","orcid":"https://orcid.org/0000-0002-8597-6795"},"institutions":[{"id":"https://openalex.org/I4210159266","display_name":"Mitsubishi Electric (United States)","ror":"https://ror.org/053jnhe44","country_code":"US","type":"company","lineage":["https://openalex.org/I1306287861","https://openalex.org/I4210133125","https://openalex.org/I4210159266"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gordon Wichern","raw_affiliation_strings":["Mitsubishi Electric Research Labs (MERL),Cambridge,MA,USA","Mitsubishi Electric Research Labs (MERL), Cambridge, MA, USA"],"affiliations":[{"raw_affiliation_string":"Mitsubishi Electric Research Labs (MERL),Cambridge,MA,USA","institution_ids":["https://openalex.org/I4210159266"]},{"raw_affiliation_string":"Mitsubishi Electric Research Labs (MERL), Cambridge, MA, USA","institution_ids":["https://openalex.org/I4210159266"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091834525","display_name":"Jee-weon Jung","orcid":"https://orcid.org/0000-0003-0505-2988"},"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":"Jee-Weon Jung","raw_affiliation_strings":["Carnegie Mellon University,Language Technologies Institute,Pittsburgh,PA,USA","Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA, USA"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University,Language Technologies Institute,Pittsburgh,PA,USA","institution_ids":["https://openalex.org/I74973139"]},{"raw_affiliation_string":"Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102812631","display_name":"Fran\u00e7ois G. Germain","orcid":"https://orcid.org/0000-0002-8973-5315"},"institutions":[{"id":"https://openalex.org/I4210159266","display_name":"Mitsubishi Electric (United States)","ror":"https://ror.org/053jnhe44","country_code":"US","type":"company","lineage":["https://openalex.org/I1306287861","https://openalex.org/I4210133125","https://openalex.org/I4210159266"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Fran\u00e7ois Germain","raw_affiliation_strings":["Mitsubishi Electric Research Labs (MERL),Cambridge,MA,USA","Mitsubishi Electric Research Labs (MERL), Cambridge, MA, USA"],"affiliations":[{"raw_affiliation_string":"Mitsubishi Electric Research Labs (MERL),Cambridge,MA,USA","institution_ids":["https://openalex.org/I4210159266"]},{"raw_affiliation_string":"Mitsubishi Electric Research Labs (MERL), Cambridge, MA, USA","institution_ids":["https://openalex.org/I4210159266"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064097430","display_name":"Jonathan Le Roux","orcid":"https://orcid.org/0000-0002-0158-2837"},"institutions":[{"id":"https://openalex.org/I4210159266","display_name":"Mitsubishi Electric (United States)","ror":"https://ror.org/053jnhe44","country_code":"US","type":"company","lineage":["https://openalex.org/I1306287861","https://openalex.org/I4210133125","https://openalex.org/I4210159266"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jonathan Le Roux","raw_affiliation_strings":["Mitsubishi Electric Research Labs (MERL),Cambridge,MA,USA","Mitsubishi Electric Research Labs (MERL), Cambridge, MA, USA"],"affiliations":[{"raw_affiliation_string":"Mitsubishi Electric Research Labs (MERL),Cambridge,MA,USA","institution_ids":["https://openalex.org/I4210159266"]},{"raw_affiliation_string":"Mitsubishi Electric Research Labs (MERL), Cambridge, MA, USA","institution_ids":["https://openalex.org/I4210159266"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001291873","display_name":"Shinji Watanabe","orcid":"https://orcid.org/0000-0002-5970-8631"},"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":"Shinji Watanabe","raw_affiliation_strings":["Carnegie Mellon University,Language Technologies Institute,Pittsburgh,PA,USA","Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA, USA"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University,Language Technologies Institute,Pittsburgh,PA,USA","institution_ids":["https://openalex.org/I74973139"]},{"raw_affiliation_string":"Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5088967620"],"corresponding_institution_ids":["https://openalex.org/I74973139"],"apc_list":null,"apc_paid":null,"fwci":4.4918,"has_fulltext":false,"cited_by_count":13,"citation_normalized_percentile":{"value":0.95166163,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"316","last_page":"320"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9998000264167786,"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":0.9998000264167786,"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":0.9950000047683716,"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/T10860","display_name":"Speech and Audio Processing","score":0.9916999936103821,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/closed-captioning","display_name":"Closed captioning","score":0.8361847996711731},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.813942551612854},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5421261787414551},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5360769033432007},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.5002660751342773},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4937227666378021},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4713040292263031},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.43757423758506775},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4199894070625305},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.1068635880947113}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.8361847996711731},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.813942551612854},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5421261787414551},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5360769033432007},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5002660751342773},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4937227666378021},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4713040292263031},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.43757423758506775},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4199894070625305},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.1068635880947113},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10447215","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447215","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":[{"id":"https://openalex.org/F4320307110","display_name":"Delta","ror":"https://ror.org/03g9c1e75"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":34,"referenced_works":["https://openalex.org/W2593116425","https://openalex.org/W2842511635","https://openalex.org/W2936774411","https://openalex.org/W2963084599","https://openalex.org/W2964213897","https://openalex.org/W3015591594","https://openalex.org/W3034999214","https://openalex.org/W3094550259","https://openalex.org/W3097777922","https://openalex.org/W3153469116","https://openalex.org/W3204358542","https://openalex.org/W3205743929","https://openalex.org/W3205860970","https://openalex.org/W3209984917","https://openalex.org/W4226442948","https://openalex.org/W4288089799","https://openalex.org/W4297808394","https://openalex.org/W4312048190","https://openalex.org/W4322718191","https://openalex.org/W4372266552","https://openalex.org/W4385570594","https://openalex.org/W4386576685","https://openalex.org/W4400033239","https://openalex.org/W6745136726","https://openalex.org/W6755207826","https://openalex.org/W6761551260","https://openalex.org/W6769627184","https://openalex.org/W6774314701","https://openalex.org/W6799494382","https://openalex.org/W6810738896","https://openalex.org/W6840200333","https://openalex.org/W6848208918","https://openalex.org/W6850625674","https://openalex.org/W6851994061"],"related_works":["https://openalex.org/W4310447809","https://openalex.org/W4200243030","https://openalex.org/W2800782462","https://openalex.org/W3209117276","https://openalex.org/W4388184981","https://openalex.org/W4323777661","https://openalex.org/W4206178588","https://openalex.org/W3094491777","https://openalex.org/W3214715529","https://openalex.org/W4287635093"],"abstract_inverted_index":{"Automated":[0],"audio":[1,81,101,137],"captioning":[2],"(AAC)":[3],"aims":[4],"to":[5,56,78,88,120,157],"generate":[6],"informative":[7],"descriptions":[8],"for":[9],"various":[10],"sounds":[11],"from":[12],"nature":[13],"and/or":[14],"human":[15],"activities.":[16],"In":[17],"recent":[18],"years,":[19],"AAC":[20,62,172,196],"has":[21,167],"quickly":[22],"attracted":[23],"research":[24],"interest,":[25],"with":[26,134],"state-of-the-art":[27,182],"systems":[28],"now":[29],"relying":[30],"on":[31,186],"a":[32,112,162,180],"sequence-to-sequence":[33],"(seq2seq)":[34],"backbone":[35],"powered":[36],"by":[37,64],"strong":[38],"models":[39,63,68,72],"such":[40],"as":[41],"Transformers.":[42],"Following":[43],"the":[44,58,135,142,146,187,193],"macro-trend":[45],"of":[46,60,92,129,150],"applied":[47],"machine":[48],"learning":[49],"research,":[50],"in":[51,171],"this":[52],"work,":[53],"we":[54,75,84,110,155],"strive":[55],"improve":[57],"performance":[59],"seq2seq":[61],"extensively":[65],"leveraging":[66],"pretrained":[67],"and":[69,94,126,148,161,191],"large":[70],"language":[71],"(LLMs).":[73],"Specifically,":[74],"utilize":[76],"BEATS":[77],"extract":[79],"fine-grained":[80],"features.":[82],"Then,":[83],"employ":[85,158],"Instructor":[86],"LLM":[87],"fetch":[89],"text":[90],"embeddings":[91],"captions,":[93],"infuse":[95],"their":[96],"language-modality":[97],"knowledge":[98],"into":[99],"BEATs":[100],"features":[102],"via":[103],"an":[104],"auxiliary":[105],"InfoNCE":[106],"loss":[107],"function.":[108],"Moreover,":[109],"propose":[111,156],"novel":[113],"data":[114],"augmentation":[115],"method":[116],"that":[117],"uses":[118],"ChatGPT":[119],"produce":[121],"caption":[122],"mix-ups":[123],"(i.e.,":[124],"grammatical":[125],"compact":[127],"combinations":[128],"two":[130],"captions)":[131],"which,":[132],"together":[133],"corresponding":[136],"mixtures,":[138],"increase":[139],"not":[140,168],"only":[141],"amount":[143],"but":[144],"also":[145],"complexity":[147],"diversity":[149],"training":[151],"data.":[152],"During":[153],"inference,":[154],"nucleus":[159],"sampling":[160],"hybrid":[163],"reranking":[164],"algorithm,":[165],"which":[166],"been":[169],"explored":[170],"research.":[173],"Combining":[174],"our":[175,177],"efforts,":[176],"model":[178],"achieves":[179],"new":[181],"32.6":[183],"SPIDEr-FL":[184],"score":[185],"Clotho":[188],"evaluation":[189],"split,":[190],"wins":[192],"2023":[194],"DCASE":[195],"challenge.":[197]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":9},{"year":2024,"cited_by_count":2}],"updated_date":"2026-04-03T22:45:19.894376","created_date":"2025-10-10T00:00:00"}