{"id":"https://openalex.org/W4415539788","doi":"https://doi.org/10.1145/3746027.3758284","title":"AudioAtlas: A Comprehensive and Balanced Benchmark Towards Movie-Oriented Text-to-Audio Generation","display_name":"AudioAtlas: A Comprehensive and Balanced Benchmark Towards Movie-Oriented Text-to-Audio Generation","publication_year":2025,"publication_date":"2025-10-25","ids":{"openalex":"https://openalex.org/W4415539788","doi":"https://doi.org/10.1145/3746027.3758284"},"language":null,"primary_location":{"id":"doi:10.1145/3746027.3758284","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746027.3758284","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","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/A5103325598","display_name":"Chenxi Wang","orcid":"https://orcid.org/0009-0003-1457-5120"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chenxi Wang","raw_affiliation_strings":["University of Science and Technology of China, Hefei, Anhui, China"],"raw_orcid":"https://orcid.org/0009-0003-1457-5120","affiliations":[{"raw_affiliation_string":"University of Science and Technology of China, Hefei, Anhui, China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102970284","display_name":"Yusheng Dai","orcid":"https://orcid.org/0000-0002-1290-2568"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yusheng Dai","raw_affiliation_strings":["University of Science and Technology of China, Hefei, Anhui, China"],"raw_orcid":"https://orcid.org/0000-0002-1290-2568","affiliations":[{"raw_affiliation_string":"University of Science and Technology of China, Hefei, Anhui, China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101926390","display_name":"Lei Sun","orcid":"https://orcid.org/0000-0001-7680-6455"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lei Sun","raw_affiliation_strings":["iFLYTEK Research, Hefei, Anhui, China"],"raw_orcid":"https://orcid.org/0000-0001-7680-6455","affiliations":[{"raw_affiliation_string":"iFLYTEK Research, Hefei, Anhui, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066595711","display_name":"Jun Du","orcid":"https://orcid.org/0000-0002-2387-0389"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Du","raw_affiliation_strings":["University of Science and Technology of China, Hefei, Anhui, China"],"raw_orcid":"https://orcid.org/0000-0002-2387-0389","affiliations":[{"raw_affiliation_string":"University of Science and Technology of China, Hefei, Anhui, China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5020540834","display_name":"Jianqing Gao","orcid":"https://orcid.org/0000-0001-5575-4940"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jianqing Gao","raw_affiliation_strings":["iFLYTEK Research, Hefei, Anhui, China"],"raw_orcid":"https://orcid.org/0000-0001-5575-4940","affiliations":[{"raw_affiliation_string":"iFLYTEK Research, Hefei, Anhui, China","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"13266","last_page":"13272"},"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/T11439","display_name":"Video Analysis and Summarization","score":0.9947999715805054,"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"}},{"id":"https://openalex.org/T11349","display_name":"Music Technology and Sound Studies","score":0.9929999709129333,"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/benchmark","display_name":"Benchmark (surveying)","score":0.6850000023841858},{"id":"https://openalex.org/keywords/categorical-variable","display_name":"Categorical variable","score":0.58160001039505},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.548799991607666},{"id":"https://openalex.org/keywords/event","display_name":"Event (particle physics)","score":0.5026000142097473},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.3230000138282776},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.2939000129699707}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8220999836921692},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6850000023841858},{"id":"https://openalex.org/C5274069","wikidata":"https://www.wikidata.org/wiki/Q2285707","display_name":"Categorical variable","level":2,"score":0.58160001039505},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.548799991607666},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.5026000142097473},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44780001044273376},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38370001316070557},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3230000138282776},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.2939000129699707},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.29319998621940613},{"id":"https://openalex.org/C2776187449","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Natural language generation","level":3,"score":0.2919999957084656},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2784999907016754}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746027.3758284","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746027.3758284","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":5,"referenced_works":["https://openalex.org/W1956340063","https://openalex.org/W2593116425","https://openalex.org/W3015591594","https://openalex.org/W4390873332","https://openalex.org/W4403780831"],"related_works":[],"abstract_inverted_index":{"Recent":[0],"rapid":[1],"progress":[2],"in":[3,13,42,129],"Text-to-Audio":[4],"(T2A)":[5],"models":[6,62],"contrasts":[7],"sharply":[8],"with":[9,92,151,173],"the":[10,14,209],"stagnation":[11],"observed":[12],"evolution":[15],"of":[16,105,115],"corresponding":[17],"evaluation":[18,55,104,114],"benchmarks.":[19],"Existing":[20],"benchmarks,":[21],"such":[22],"as":[23,31,33],"AudioCaps,":[24],"suffer":[25],"from":[26,99],"limited":[27],"diversity":[28],"and":[29,53,83,89,108,135,175,187,202],"quality,":[30],"well":[32],"biased":[34],"category":[35,72],"distributions,":[36],"leading":[37],"to":[38,197],"increasingly":[39],"questionable":[40],"reliability":[41],"assessing":[43],"advanced":[44],"T2A":[45,61,144,160,171],"models.":[46,138],"This":[47],"paper":[48],"introduces":[49],"AudioAtlas,":[50],"a":[51,165],"comprehensive":[52,166],"balanced":[54],"benchmark":[56],"specifically":[57,195],"designed":[58],"for":[59,158,191],"evaluating":[60],"aimed":[63],"at":[64],"movie":[65],"production.":[66],"Based":[67],"on":[68,169,208],"an":[69],"object-centric":[70],"audio":[71],"system,":[73],"AudioAtlas":[74,146],"provides":[75,147],"high-quality":[76],"reference":[77],"samples":[78],"characterized":[79],"by":[80,180],"categorical":[81],"balance":[82],"diversity.":[84],"It":[85],"includes":[86],"detailed":[87],"overall":[88],"event-level":[90],"captions":[91],"rich":[93],"descriptors,":[94],"plus":[95],"fine-grained":[96],"temporal":[97,106],"annotations":[98],"human":[100],"experts,":[101],"enabling":[102],"thorough":[103],"alignment":[107],"semantic":[109],"accuracy.":[110],"To":[111],"enable":[112],"precise":[113],"temporally-aligned":[116],"generation":[117],"across":[118],"universal":[119],"categories,":[120],"two":[121],"novel":[122],"metrics":[123],"are":[124,206],"proposed":[125],"leveraging":[126],"recent":[127],"advancements":[128],"large-scale":[130],"Audio":[131,194],"Language":[132],"Models":[133],"(AudioLLMs)":[134],"contrastive":[136],"learning":[137],"By":[139],"re-benchmarking":[140],"six":[141],"currently":[142],"influential":[143],"models,":[145],"evaluations":[148],"better":[149],"aligned":[150],"aesthetic":[152],"considerations,":[153],"offering":[154],"clearer":[155],"optimization":[156],"directions":[157,190],"movie-production-oriented":[159],"systems.":[161],"Additionally,":[162],"we":[163],"conduct":[164],"comparative":[167],"analysis":[168],"temporally-controllable":[170],"methods":[172],"training-based,":[174],"promising":[176],"training-free":[177],"approaches":[178],"inspired":[179],"region-controllable":[181],"image":[182],"generation,":[183],"clarifying":[184],"current":[185],"limitations":[186],"pointing":[188],"out":[189],"future":[192],"research.":[193],"refers":[196],"sound":[198],"event":[199],"excluding":[200],"speech":[201],"music.":[203],"Further":[204],"details":[205],"available":[207],"project":[210],"page:":[211],"https://audioatlas.github.io/AudioAtlas/":[212]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-25T00:00:00"}