{"id":"https://openalex.org/W4392909532","doi":"https://doi.org/10.1109/icassp48485.2024.10446973","title":"SECP: A Speech Enhancement-Based Curation Pipeline for Scalable Acquisition of Clean Speech","display_name":"SECP: A Speech Enhancement-Based Curation Pipeline for Scalable Acquisition of Clean Speech","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392909532","doi":"https://doi.org/10.1109/icassp48485.2024.10446973"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10446973","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp48485.2024.10446973","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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/A5093975963","display_name":"Adam Sabra","orcid":null},"institutions":[{"id":"https://openalex.org/I135428043","display_name":"Cisco Systems (United States)","ror":"https://ror.org/03yt1ez60","country_code":"US","type":"company","lineage":["https://openalex.org/I135428043"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Adam Sabra","raw_affiliation_strings":["Cisco Systems, Inc"],"affiliations":[{"raw_affiliation_string":"Cisco Systems, Inc","institution_ids":["https://openalex.org/I135428043"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074271937","display_name":"Cyprian Wronka","orcid":null},"institutions":[{"id":"https://openalex.org/I135428043","display_name":"Cisco Systems (United States)","ror":"https://ror.org/03yt1ez60","country_code":"US","type":"company","lineage":["https://openalex.org/I135428043"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Cyprian Wronka","raw_affiliation_strings":["Cisco Systems, Inc"],"affiliations":[{"raw_affiliation_string":"Cisco Systems, Inc","institution_ids":["https://openalex.org/I135428043"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101319291","display_name":"Michelle Mao","orcid":null},"institutions":[{"id":"https://openalex.org/I135428043","display_name":"Cisco Systems (United States)","ror":"https://ror.org/03yt1ez60","country_code":"US","type":"company","lineage":["https://openalex.org/I135428043"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Michelle Mao","raw_affiliation_strings":["Cisco Systems, Inc"],"affiliations":[{"raw_affiliation_string":"Cisco Systems, Inc","institution_ids":["https://openalex.org/I135428043"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5093975964","display_name":"Samer Hijazi","orcid":null},"institutions":[{"id":"https://openalex.org/I135428043","display_name":"Cisco Systems (United States)","ror":"https://ror.org/03yt1ez60","country_code":"US","type":"company","lineage":["https://openalex.org/I135428043"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Samer Hijazi","raw_affiliation_strings":["Cisco Systems, Inc"],"affiliations":[{"raw_affiliation_string":"Cisco Systems, Inc","institution_ids":["https://openalex.org/I135428043"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5093975963"],"corresponding_institution_ids":["https://openalex.org/I135428043"],"apc_list":null,"apc_paid":null,"fwci":0.3626,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.61709036,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"11981","last_page":"11985"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9869999885559082,"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"}},"topics":[{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9869999885559082,"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/T10860","display_name":"Speech and Audio Processing","score":0.9800000190734863,"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/T12031","display_name":"Speech and dialogue systems","score":0.9746999740600586,"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/pipeline","display_name":"Pipeline (software)","score":0.7931065559387207},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.709147572517395},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6891405582427979},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.47244173288345337},{"id":"https://openalex.org/keywords/voice-activity-detection","display_name":"Voice activity detection","score":0.4208931624889374},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.3626193404197693},{"id":"https://openalex.org/keywords/database","display_name":"Database","score":0.10368460416793823},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.0953809916973114}],"concepts":[{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.7931065559387207},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.709147572517395},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6891405582427979},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.47244173288345337},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.4208931624889374},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.3626193404197693},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.10368460416793823},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0953809916973114}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10446973","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp48485.2024.10446973","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":18,"referenced_works":["https://openalex.org/W1552314771","https://openalex.org/W1901129140","https://openalex.org/W2094721231","https://openalex.org/W2095629250","https://openalex.org/W2113577323","https://openalex.org/W2507424718","https://openalex.org/W2526733715","https://openalex.org/W2963453742","https://openalex.org/W2972359262","https://openalex.org/W2973133192","https://openalex.org/W2973196014","https://openalex.org/W3193786050","https://openalex.org/W3198020407","https://openalex.org/W4205371873","https://openalex.org/W4300505472","https://openalex.org/W6630864721","https://openalex.org/W6738603831","https://openalex.org/W6917585676"],"related_works":["https://openalex.org/W3135230428","https://openalex.org/W2904739811","https://openalex.org/W3087422378","https://openalex.org/W2152158029","https://openalex.org/W2012540220","https://openalex.org/W2559837139","https://openalex.org/W1151175420","https://openalex.org/W2131711534","https://openalex.org/W2559040841","https://openalex.org/W114661351"],"abstract_inverted_index":{"As":[0],"more":[1],"speech":[2,13,23,72,77],"technologies":[3],"rely":[4],"on":[5,36],"a":[6,18,43,64,76,117],"supervised":[7],"deep":[8],"learning":[9],"approach":[10,30],"with":[11],"clean":[12,68,71],"as":[14,63,104],"the":[15,34,84,90,136,148],"ground":[16,105],"truth,":[17],"methodology":[19],"to":[20,32,66,113],"onboard":[21,67],"said":[22],"at":[24],"scale":[25],"is":[26,144],"needed.":[27,46],"However,":[28],"this":[29,48,52,121],"needs":[31],"minimize":[33],"dependency":[35],"human":[37],"listening":[38],"and":[39,87,138],"annotation,":[40],"only":[41],"requiring":[42],"human-in-the-loop":[44],"when":[45],"In":[47],"paper,":[49],"we":[50,98],"address":[51],"issue":[53],"by":[54],"outlining":[55],"Speech":[56],"Enhancement-based":[57],"Curation":[58],"Pipeline":[59],"(SECP)":[60],"which":[61,80],"serves":[62],"framework":[65],"speech.":[69],"This":[70],"can":[73,81],"then":[74],"train":[75],"enhancement":[78],"model,":[79],"further":[82],"refine":[83],"original":[85,149],"dataset":[86],"thus":[88],"close":[89],"iterative":[91,96],"loop.":[92],"By":[93],"running":[94],"two":[95],"rounds,":[97],"observe":[99],"that":[100,135],"enhanced":[101],"output":[102],"used":[103,119],"truth":[106],"does":[107],"not":[108],"degrade":[109],"model":[110],"performance":[111],"according":[112],"\u2206<inf":[114],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[115],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">PESQ</inf>,":[116],"metric":[118],"in":[120],"paper.":[122],"We":[123],"also":[124],"show":[125],"through":[126],"comparative":[127],"mean":[128],"opinion":[129],"score":[130],"(CMOS)":[131],"based":[132],"subjective":[133],"tests":[134],"highest":[137],"lowest":[139],"bound":[140],"of":[141],"refined":[142],"data":[143],"perceptually":[145],"better":[146],"than":[147],"data.":[150]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2025-12-22T23:10:17.713674","created_date":"2025-10-10T00:00:00"}