{"id":"https://openalex.org/W4408352769","doi":"https://doi.org/10.1109/icassp49660.2025.10888074","title":"Elevating Robust ASR By Decoupling Multi-Channel Speaker Separation and Speech Recognition","display_name":"Elevating Robust ASR By Decoupling Multi-Channel Speaker Separation and Speech Recognition","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408352769","doi":"https://doi.org/10.1109/icassp49660.2025.10888074"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10888074","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888074","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":"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/A5100730499","display_name":"Yufeng Yang","orcid":"https://orcid.org/0000-0003-0780-2703"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yufeng Yang","raw_affiliation_strings":["The Ohio State University,Department of Computer Science and Engineering,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University,Department of Computer Science and Engineering,USA","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073861062","display_name":"Hassan Taherian","orcid":"https://orcid.org/0000-0002-7548-3081"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hassan Taherian","raw_affiliation_strings":["The Ohio State University,Department of Computer Science and Engineering,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University,Department of Computer Science and Engineering,USA","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009379093","display_name":"Vahid Ahmadi Kalkhorani","orcid":"https://orcid.org/0000-0002-4166-1935"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Vahid Ahmadi Kalkhorani","raw_affiliation_strings":["The Ohio State University,Department of Computer Science and Engineering,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University,Department of Computer Science and Engineering,USA","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051837453","display_name":"DeLiang Wang","orcid":"https://orcid.org/0000-0001-8195-6319"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"DeLiang Wang","raw_affiliation_strings":["The Ohio State University,Department of Computer Science and Engineering,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University,Department of Computer Science and Engineering,USA","institution_ids":["https://openalex.org/I52357470"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I52357470"],"apc_list":null,"apc_paid":null,"fwci":1.0139,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.7135648,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":97},"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":0.9980999827384949,"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":0.9980999827384949,"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":0.9922000169754028,"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/T11447","display_name":"Blind Source Separation Techniques","score":0.9577000141143799,"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/speech-recognition","display_name":"Speech recognition","score":0.8058489561080933},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7299388647079468},{"id":"https://openalex.org/keywords/decoupling","display_name":"Decoupling (probability)","score":0.6174858212471008},{"id":"https://openalex.org/keywords/speaker-recognition","display_name":"Speaker recognition","score":0.6128591895103455},{"id":"https://openalex.org/keywords/channel","display_name":"Channel (broadcasting)","score":0.44644734263420105},{"id":"https://openalex.org/keywords/speaker-verification","display_name":"Speaker verification","score":0.4356621205806732},{"id":"https://openalex.org/keywords/separation","display_name":"Separation (statistics)","score":0.4345664978027344},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.36992037296295166},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1228070855140686},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.09263491630554199},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.06897518038749695},{"id":"https://openalex.org/keywords/control-engineering","display_name":"Control engineering","score":0.06888836622238159}],"concepts":[{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.8058489561080933},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7299388647079468},{"id":"https://openalex.org/C205606062","wikidata":"https://www.wikidata.org/wiki/Q5249645","display_name":"Decoupling (probability)","level":2,"score":0.6174858212471008},{"id":"https://openalex.org/C133892786","wikidata":"https://www.wikidata.org/wiki/Q1145189","display_name":"Speaker recognition","level":2,"score":0.6128591895103455},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.44644734263420105},{"id":"https://openalex.org/C2982762665","wikidata":"https://www.wikidata.org/wiki/Q1145189","display_name":"Speaker verification","level":3,"score":0.4356621205806732},{"id":"https://openalex.org/C2776061190","wikidata":"https://www.wikidata.org/wiki/Q7451805","display_name":"Separation (statistics)","level":2,"score":0.4345664978027344},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36992037296295166},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1228070855140686},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.09263491630554199},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.06897518038749695},{"id":"https://openalex.org/C133731056","wikidata":"https://www.wikidata.org/wiki/Q4917288","display_name":"Control engineering","level":1,"score":0.06888836622238159}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10888074","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888074","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":[{"display_name":"Reduced inequalities","score":0.5,"id":"https://metadata.un.org/sdg/10"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320307110","display_name":"Delta","ror":"https://ror.org/03g9c1e75"},{"id":"https://openalex.org/F4320317189","display_name":"Ohio Supercomputer Center","ror":"https://ror.org/01apna436"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":43,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1552314771","https://openalex.org/W2024490156","https://openalex.org/W2101045344","https://openalex.org/W2125838338","https://openalex.org/W2141998673","https://openalex.org/W2160815625","https://openalex.org/W2291877678","https://openalex.org/W2559260703","https://openalex.org/W2734774145","https://openalex.org/W2952218014","https://openalex.org/W2962780374","https://openalex.org/W2962866211","https://openalex.org/W2963843276","https://openalex.org/W2980451698","https://openalex.org/W3015372568","https://openalex.org/W3016094953","https://openalex.org/W3016232124","https://openalex.org/W3020336359","https://openalex.org/W3032514799","https://openalex.org/W3094821064","https://openalex.org/W3094831814","https://openalex.org/W3161590371","https://openalex.org/W3165776651","https://openalex.org/W3171278394","https://openalex.org/W3182358018","https://openalex.org/W3204618061","https://openalex.org/W3209984917","https://openalex.org/W4226390724","https://openalex.org/W4283022709","https://openalex.org/W4294891356","https://openalex.org/W4385756463","https://openalex.org/W4385823058","https://openalex.org/W4386763469","https://openalex.org/W4388017359","https://openalex.org/W4390660238","https://openalex.org/W4391621214","https://openalex.org/W4395471098","https://openalex.org/W4404102497","https://openalex.org/W4410529749","https://openalex.org/W6631362777","https://openalex.org/W6766475400","https://openalex.org/W6769802879"],"related_works":["https://openalex.org/W66821593","https://openalex.org/W1521299571","https://openalex.org/W2791440692","https://openalex.org/W3141593045","https://openalex.org/W204267554","https://openalex.org/W2134501921","https://openalex.org/W4252590334","https://openalex.org/W2543777506","https://openalex.org/W2407001684","https://openalex.org/W3096066489"],"abstract_inverted_index":{"Despite":[0],"the":[1,10,42,71,74,80,84,93,105,116,122,128],"tremendous":[2],"success":[3],"of":[4,12,73,102,119],"automatic":[5],"speech":[6,59,139],"recognition":[7,142],"(ASR)":[8],"with":[9,83],"introduction":[11],"deep":[13],"learning,":[14],"its":[15],"performance":[16],"is":[17,143],"still":[18],"unsatisfactory":[19],"in":[20,28],"many":[21],"real-world":[22],"multi-talker":[23],"scenarios.":[24],"Speaker":[25],"separation":[26,77,140],"excels":[27],"separating":[29],"individual":[30],"talkers":[31],"but,":[32],"as":[33],"a":[34,50,97,144],"frontend,":[35],"it":[36],"introduces":[37],"processing":[38,62],"artifacts":[39],"that":[40,137],"degrade":[41],"ASR":[43,54,81],"backend":[44],"trained":[45,86,126],"on":[46,57,88,111,127],"clean":[47,89],"speech.":[48,90],"As":[49],"result,":[51],"mainstream":[52],"robust":[53,149],"systems":[55],"train":[56],"noisy":[58],"to":[60,69,148],"avoid":[61],"artifacts.":[63],"In":[64],"this":[65],"work,":[66],"we":[67,114],"propose":[68],"decouple":[70],"training":[72],"multi-channel":[75],"speaker":[76],"frontend":[78],"and":[79,141],"backend,":[82],"latter":[85],"only":[87],"On":[91],"SMS-WSJ,":[92],"proposed":[94],"approach":[95,147],"achieves":[96],"word":[98],"error":[99],"rate":[100],"(WER)":[101],"5.74%,":[103],"outperforming":[104,121],"previous":[106,123],"best":[107,124],"by":[108,131],"14.3%.":[109],"Furthermore,":[110],"recorded":[112],"LibriCSS,":[113],"achieve":[115],"speaker-attributed":[117],"WER":[118],"3.86%,":[120],"system":[125],"same":[129],"data":[130],"24.8%.":[132],"These":[133],"state-of-the-art":[134],"results":[135],"suggest":[136],"decoupling":[138],"potentially":[145],"effective":[146],"ASR.":[150]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}