{"id":"https://openalex.org/W4408354646","doi":"https://doi.org/10.1109/icassp49660.2025.10887588","title":"An Efficient and Streaming Audio Visual Active Speaker Detection System","display_name":"An Efficient and Streaming Audio Visual Active Speaker Detection System","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408354646","doi":"https://doi.org/10.1109/icassp49660.2025.10887588"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10887588","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10887588","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 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/A5034953394","display_name":"Arnav Kundu","orcid":"https://orcid.org/0000-0003-2921-4312"},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Arnav Kundu","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008400874","display_name":"Yanzi Jin","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yanzi Jin","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5095886473","display_name":"Mohammad Hossein Sekhavat","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Mohammad Sekhavat","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012428670","display_name":"Maxwell Horton","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Maxwell Horton","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114367900","display_name":"Danny Tormoen","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Danny Tormoen","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5113880378","display_name":"Devang Naik","orcid":"https://orcid.org/0009-0007-7838-1623"},"institutions":[{"id":"https://openalex.org/I4210107260","display_name":"Apple (United Kingdom)","ror":"https://ror.org/01vpeym60","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210107260"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Devang Naik","raw_affiliation_strings":["Apple"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple","institution_ids":["https://openalex.org/I4210107260"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210107260"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.04366035,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9965999722480774,"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/T10860","display_name":"Speech and Audio Processing","score":0.9965999722480774,"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/T11447","display_name":"Blind Source Separation Techniques","score":0.95660001039505,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.9520000219345093,"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":0.8172170519828796},{"id":"https://openalex.org/keywords/audio-visual","display_name":"Audio visual","score":0.5779085755348206},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5674830079078674},{"id":"https://openalex.org/keywords/video-streaming","display_name":"Video streaming","score":0.45236676931381226},{"id":"https://openalex.org/keywords/multimedia","display_name":"Multimedia","score":0.2873796224594116},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.26186108589172363}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8172170519828796},{"id":"https://openalex.org/C3017588708","wikidata":"https://www.wikidata.org/wiki/Q758901","display_name":"Audio visual","level":2,"score":0.5779085755348206},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5674830079078674},{"id":"https://openalex.org/C2986160907","wikidata":"https://www.wikidata.org/wiki/Q220499","display_name":"Video streaming","level":2,"score":0.45236676931381226},{"id":"https://openalex.org/C49774154","wikidata":"https://www.wikidata.org/wiki/Q131765","display_name":"Multimedia","level":1,"score":0.2873796224594116},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.26186108589172363}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10887588","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10887588","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.550000011920929,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W2330149154","https://openalex.org/W2963155035","https://openalex.org/W3016098309","https://openalex.org/W3034702511","https://openalex.org/W3206008172","https://openalex.org/W4224923231","https://openalex.org/W4312268709","https://openalex.org/W4386075783","https://openalex.org/W4392908938","https://openalex.org/W4402754289","https://openalex.org/W6631190155","https://openalex.org/W6726497184","https://openalex.org/W6739901393","https://openalex.org/W6846659131"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2271369634","https://openalex.org/W3147472394","https://openalex.org/W2047100085","https://openalex.org/W2350550760","https://openalex.org/W578794879","https://openalex.org/W2293038204","https://openalex.org/W3082618691"],"abstract_inverted_index":{"This":[0,156],"paper":[1],"delves":[2],"into":[3],"the":[4,13,55,87,101,109,117,121,144,150,158,216,253,263,271],"challenging":[5],"task":[6],"of":[7,30,57,103,124,147,208,220,255],"active":[8],"speaker":[9],"detection":[10],"(asd),":[11],"where":[12],"system":[14,59],"needs":[15],"to":[16,99,176,191,270],"determine":[17],"in":[18,27,40,54],"real-time":[19,58,92],"whether":[20],"a":[21,28,50,97,128,138,204,229,241],"person":[22],"is":[23,130,249,266],"speaking":[24],"or":[25,192],"not":[26],"series":[29],"video":[31],"frames.":[32,210],"While":[33],"previous":[34],"works":[35],"have":[36],"made":[37],"significant":[38],"strides":[39],"improving":[41],"network":[42],"architectures":[43],"and":[44,68,261],"learning":[45],"effective":[46],"representations":[47],"for":[48,74,119],"ASD,":[49],"critical":[51],"gap":[52],"exists":[53],"exploration":[56],"deployment.":[60],"Existing":[61],"models":[62,186],"often":[63],"suffer":[64],"from":[65],"high":[66],"latency":[67],"memory":[69,160,218,250,272],"usage,":[70],"rendering":[71],"them":[72],"impractical":[73],"immediate":[75],"applications.":[76],"To":[77],"bridge":[78],"this":[79],"gap,":[80],"we":[81,95,115,136,172,212,243],"present":[82],"two":[83],"scenarios":[84],"that":[85,142,183,224,245,262],"address":[86],"key":[88],"challenges":[89],"posed":[90],"by":[91,108,252],"constraints.":[93],"First,":[94],"introduce":[96],"method":[98],"limit":[100],"number":[102,146,207],"future":[104,125,236],"context":[105,209,227,257],"frames":[106,126,149],"utilized":[107],"ASD":[110,166,221],"model.":[111],"By":[112],"doing":[113],"so,":[114],"alleviate":[116],"need":[118],"processing":[120],"entire":[122],"sequence":[123],"before":[127],"decision":[129],"made,":[131],"significantly":[132,205],"reducing":[133],"latency.":[134],"Second,":[135],"propose":[137],"more":[139,230],"stringent":[140],"constraint":[141],"limits":[143],"total":[145],"past":[148,226,256],"model":[151],"can":[152,187,259],"access":[153],"during":[154],"inference.":[155],"tackles":[157],"persistent":[159],"issues":[161],"associated":[162],"with":[163,203],"running":[164],"streaming":[165],"systems.":[167],"Beyond":[168],"these":[169],"theoretical":[170],"frameworks,":[171],"conduct":[173],"extensive":[174],"experiments":[175],"validate":[177],"our":[178,246],"approach.":[179],"Our":[180],"results":[181],"demonstrate":[182],"constrained":[184],"transformer":[185],"achieve":[188],"performance":[189],"comparable":[190],"even":[193],"better":[194],"than":[195,235],"state-of-the-art":[196],"recurrent":[197],"models,":[198],"such":[199],"as":[200,268],"uni-directional":[201],"GRUs,":[202],"reduced":[206],"Moreover,":[211],"shed":[213],"light":[214],"on":[215,233,240],"temporal":[217],"requirements":[219],"systems,":[222],"revealing":[223],"larger":[225],"has":[228],"profound":[231],"impact":[232],"accuracy":[234],"context.":[237],"When":[238],"profiling":[239],"CPU":[242],"find":[244],"efficient":[247],"architecture":[248],"bound":[251],"amount":[254],"it":[258],"use":[260],"compute":[264],"cost":[265],"negligible":[267],"compared":[269],"cost.":[273]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}