{"id":"https://openalex.org/W3160936850","doi":"https://doi.org/10.1109/icassp39728.2021.9413933","title":"Don\u2019t Shoot Butterfly with Rifles: Multi-Channel Continuous Speech Separation with Early Exit Transformer","display_name":"Don\u2019t Shoot Butterfly with Rifles: Multi-Channel Continuous Speech Separation with Early Exit Transformer","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3160936850","doi":"https://doi.org/10.1109/icassp39728.2021.9413933","mag":"3160936850"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9413933","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9413933","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 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/A5079533447","display_name":"Sanyuan Chen","orcid":"https://orcid.org/0000-0002-3082-6052"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Sanyuan Chen","raw_affiliation_strings":["Harbin Institute of Technology"],"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101709477","display_name":"Yu Wu","orcid":"https://orcid.org/0000-0002-5715-3011"},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Yu Wu","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108961732","display_name":"Zhuo Chen","orcid":"https://orcid.org/0009-0007-3882-3810"},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Zhuo Chen","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101618071","display_name":"Takuya Yoshioka","orcid":"https://orcid.org/0009-0003-7791-3545"},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Takuya Yoshioka","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101635405","display_name":"Shujie Liu","orcid":"https://orcid.org/0009-0008-0785-8882"},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Shujie Liu","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100365053","display_name":"Jinyu Li","orcid":"https://orcid.org/0000-0002-1089-9748"},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Jinyu Li","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5072540013","display_name":"Xiangzhan Yu","orcid":"https://orcid.org/0000-0002-1183-2844"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiangzhan Yu","raw_affiliation_strings":["Harbin Institute of Technology"],"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology","institution_ids":["https://openalex.org/I204983213"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5079533447"],"corresponding_institution_ids":["https://openalex.org/I204983213"],"apc_list":null,"apc_paid":null,"fwci":2.2854,"has_fulltext":false,"cited_by_count":19,"citation_normalized_percentile":{"value":0.88837052,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"6139","last_page":"6143"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":1.0,"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":1.0,"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/T11309","display_name":"Music and Audio Processing","score":0.9994999766349792,"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":0.9983000159263611,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7574388384819031},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.7064123749732971},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6909337043762207},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5354104042053223},{"id":"https://openalex.org/keywords/conversation","display_name":"Conversation","score":0.5092800855636597},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4364873766899109},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.12462732195854187}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7574388384819031},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7064123749732971},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6909337043762207},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5354104042053223},{"id":"https://openalex.org/C2777200299","wikidata":"https://www.wikidata.org/wiki/Q52943","display_name":"Conversation","level":2,"score":0.5092800855636597},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4364873766899109},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.12462732195854187},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9413933","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9413933","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":42,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W2030486566","https://openalex.org/W2069681747","https://openalex.org/W2221409856","https://openalex.org/W2460742184","https://openalex.org/W2734774145","https://openalex.org/W2765425905","https://openalex.org/W2803322398","https://openalex.org/W2892009249","https://openalex.org/W2908510526","https://openalex.org/W2939246703","https://openalex.org/W2944701285","https://openalex.org/W2951244744","https://openalex.org/W2952218014","https://openalex.org/W2962715207","https://openalex.org/W2963403868","https://openalex.org/W2963925437","https://openalex.org/W2964121744","https://openalex.org/W2972451902","https://openalex.org/W2972818416","https://openalex.org/W2992632249","https://openalex.org/W3001498117","https://openalex.org/W3007328579","https://openalex.org/W3008283340","https://openalex.org/W3015199127","https://openalex.org/W3015834770","https://openalex.org/W3016232124","https://openalex.org/W3095189764","https://openalex.org/W3095311338","https://openalex.org/W3099330747","https://openalex.org/W3163842642","https://openalex.org/W4385245566","https://openalex.org/W6631190155","https://openalex.org/W6739901393","https://openalex.org/W6754299077","https://openalex.org/W6757817989","https://openalex.org/W6762100871","https://openalex.org/W6768815455","https://openalex.org/W6770812743","https://openalex.org/W6772886038","https://openalex.org/W6779264309","https://openalex.org/W6782142667"],"related_works":["https://openalex.org/W1968552888","https://openalex.org/W1527532029","https://openalex.org/W2374116601","https://openalex.org/W3093134843","https://openalex.org/W2378167147","https://openalex.org/W3210777354","https://openalex.org/W4206178588","https://openalex.org/W4287635093","https://openalex.org/W3094491777","https://openalex.org/W3214715529"],"abstract_inverted_index":{"With":[0],"its":[1],"strong":[2],"modeling":[3],"capacity":[4],"that":[5,85,140],"comes":[6],"from":[7],"a":[8,15,21,43,67,77,87,97],"multi-head":[9],"and":[10,24],"multi-layer":[11],"structure,":[12],"Transformer":[13,128],"is":[14],"very":[16,88],"powerful":[17],"model":[18,129],"for":[19,46,92],"learning":[20],"sequential":[22],"representation":[23],"has":[25],"been":[26],"successfully":[27],"applied":[28],"to":[29,76,130],"speech":[30,35],"separation":[31,36,73,112],"recently.":[32],"However,":[33],"multi-channel":[34],"sometimes":[37],"does":[38,143],"not":[39,101,141],"necessarily":[40],"need":[41],"such":[42],"heavy":[44],"structure":[45,91],"all":[47],"time":[48],"frames":[49],"especially":[50],"when":[51],"the":[52,72,105,111,127,144,149,155],"cross-talker":[53],"challenge":[54],"happens":[55],"only":[56,66,102,142],"occasionally.":[57],"For":[58],"example,":[59],"in":[60],"conversation":[61],"scenarios,":[62],"most":[63],"regions":[64],"contain":[65],"single":[68,78],"active":[69],"speaker,":[70],"where":[71],"task":[74],"downgrades":[75],"speaker":[79],"enhancement":[80],"problem.":[81],"It":[82],"turns":[83],"out":[84],"using":[86],"deep":[89],"network":[90],"dealing":[93],"with":[94,96,116,134],"signals":[95],"low":[98],"overlap":[99],"ratio":[100],"negatively":[103],"affects":[104],"inference":[106],"efficiency":[107],"but":[108,151],"also":[109,153],"hurts":[110],"performance.":[113],"To":[114],"deal":[115],"this":[117],"problem,":[118],"we":[119],"propose":[120],"an":[121],"early":[122,145],"exit":[123,146],"mechanism,":[124],"which":[125],"enables":[126],"handle":[131],"different":[132],"cases":[133],"adaptive":[135],"depth.":[136],"Experimental":[137],"results":[138],"indicate":[139],"mechanism":[147],"accelerate":[148],"inference,":[150],"it":[152],"improves":[154],"accuracy.":[156]},"counts_by_year":[{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":5},{"year":2021,"cited_by_count":4}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}