{"id":"https://openalex.org/W4408355270","doi":"https://doi.org/10.1109/icassp49660.2025.10889243","title":"Multi-modal Streaming ASR in Cross-talk Scenario for Smart Glasses","display_name":"Multi-modal Streaming ASR in Cross-talk Scenario for Smart Glasses","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408355270","doi":"https://doi.org/10.1109/icassp49660.2025.10889243"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10889243","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889243","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/A5038529708","display_name":"Ya Jiang","orcid":"https://orcid.org/0000-0003-1733-8887"},"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":"Ya Jiang","raw_affiliation_strings":["University of Science and Technology of China (USTC),NERC-SLIP,Hefei,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China (USTC),NERC-SLIP,Hefei,China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101826596","display_name":"Hongbo Lan","orcid":"https://orcid.org/0000-0002-0959-5413"},"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":"Hongbo Lan","raw_affiliation_strings":["University of Science and Technology of China (USTC),NERC-SLIP,Hefei,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China (USTC),NERC-SLIP,Hefei,China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100434900","display_name":"Qing Wang","orcid":"https://orcid.org/0000-0003-3843-3920"},"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":"Qing Wang","raw_affiliation_strings":["University of Science and Technology of China (USTC),NERC-SLIP,Hefei,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China (USTC),NERC-SLIP,Hefei,China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5064699619","display_name":"Shutong Niu","orcid":"https://orcid.org/0000-0003-0315-1042"},"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":"Shutong Niu","raw_affiliation_strings":["University of Science and Technology of China (USTC),NERC-SLIP,Hefei,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China (USTC),NERC-SLIP,Hefei,China","institution_ids":["https://openalex.org/I126520041"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I126520041"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"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/T10888","display_name":"Augmented Reality Applications","score":0.8167999982833862,"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"}},"topics":[{"id":"https://openalex.org/T10888","display_name":"Augmented Reality Applications","score":0.8167999982833862,"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/T10789","display_name":"Interactive and Immersive Displays","score":0.7906000018119812,"subfield":{"id":"https://openalex.org/subfields/1709","display_name":"Human-Computer Interaction"},"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/T12784","display_name":"Modular Robots and Swarm Intelligence","score":0.7330999970436096,"subfield":{"id":"https://openalex.org/subfields/2210","display_name":"Mechanical Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6872336864471436},{"id":"https://openalex.org/keywords/modal","display_name":"Modal","score":0.6837320327758789},{"id":"https://openalex.org/keywords/materials-science","display_name":"Materials science","score":0.07586154341697693},{"id":"https://openalex.org/keywords/composite-material","display_name":"Composite material","score":0.060522109270095825}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6872336864471436},{"id":"https://openalex.org/C71139939","wikidata":"https://www.wikidata.org/wiki/Q910194","display_name":"Modal","level":2,"score":0.6837320327758789},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.07586154341697693},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.060522109270095825}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10889243","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889243","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":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":36,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1971243715","https://openalex.org/W2073957688","https://openalex.org/W2127141656","https://openalex.org/W2144499799","https://openalex.org/W2556114683","https://openalex.org/W2892009249","https://openalex.org/W2936123380","https://openalex.org/W2962742956","https://openalex.org/W3007227084","https://openalex.org/W3007498683","https://openalex.org/W3015671919","https://openalex.org/W3090160518","https://openalex.org/W3092557781","https://openalex.org/W3149509723","https://openalex.org/W3162293946","https://openalex.org/W4226491018","https://openalex.org/W4283204589","https://openalex.org/W4284883478","https://openalex.org/W4319862683","https://openalex.org/W4385245566","https://openalex.org/W4385822839","https://openalex.org/W4385965602","https://openalex.org/W4391021542","https://openalex.org/W4392902568","https://openalex.org/W4392908343","https://openalex.org/W4392910324","https://openalex.org/W4408355612","https://openalex.org/W6623517193","https://openalex.org/W6631190155","https://openalex.org/W6638667902","https://openalex.org/W6747158283","https://openalex.org/W6755207826","https://openalex.org/W6780226713","https://openalex.org/W6810168380","https://openalex.org/W6856127113"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"In":[0],"the":[1,5,22,42,48,60,66,78,85,90,116,125,131,135],"MMCSG":[2,126],"task":[3],"of":[4,50,137],"CHiME-8":[6],"Challenge,":[7],"achieving":[8,102],"real-time":[9,104],"speaker-attributed":[10],"transcriptions":[11],"with":[12,21,81,108],"limited":[13],"multi-modal":[14,35],"data":[15,80,83,96,140],"presents":[16],"significant":[17],"challenges.":[18],"To":[19],"cope":[20],"problem,":[23],"we":[24,53,76],"propose":[25],"a":[26,38,55],"novel":[27],"ASR":[28],"framework":[29],"that":[30,89],"leverages":[31],"both":[32],"audio-only":[33,43],"and":[34,46,72,93],"inputs":[36],"in":[37,84,101,124],"streaming":[39],"fashion.":[40],"For":[41],"modality,":[44],"analyzing":[45],"emulating":[47],"characteristics":[49],"real":[51,71],"audio,":[52],"utilize":[54],"multi-channel":[56],"simulation":[57],"to":[58],"generate":[59],"augmented":[61],"dataset,":[62],"which":[63],"efficiently":[64],"reduces":[65],"model":[67],"training":[68],"deviation":[69],"between":[70],"simulated":[73],"data.":[74],"Additionally,":[75],"integrate":[77],"IMU":[79,95,139],"audio":[82,99],"network":[86],"structure,":[87],"demonstrating":[88],"functional":[91],"filtered":[92],"encoded":[94],"can":[97],"assist":[98],"information":[100],"better":[103],"speech":[105],"recognition":[106],"performance":[107],"ablation":[109],"experiments.":[110],"Notably,":[111],"our":[112],"explorations":[113],"based":[114],"on":[115],"above":[117],"schemes":[118],"not":[119],"only":[120],"secured":[121],"first":[122,132],"place":[123],"sub-track":[127],"but":[128],"also":[129],"represented":[130],"investigation":[133],"into":[134],"effectiveness":[136],"leveraging":[138],"for":[141],"this":[142],"task.":[143]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}