{"id":"https://openalex.org/W4225263883","doi":"https://doi.org/10.1109/icassp43922.2022.9747036","title":"A Bayesian Permutation Training Deep Representation Learning Method for Speech Enhancement with Variational Autoencoder","display_name":"A Bayesian Permutation Training Deep Representation Learning Method for Speech Enhancement with Variational Autoencoder","publication_year":2022,"publication_date":"2022-04-27","ids":{"openalex":"https://openalex.org/W4225263883","doi":"https://doi.org/10.1109/icassp43922.2022.9747036"},"language":"en","primary_location":{"id":"doi:10.1109/icassp43922.2022.9747036","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9747036","pdf_url":null,"source":{"id":"https://openalex.org/S4363607702","display_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"http://arxiv.org/pdf/2201.09875.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5029253113","display_name":"Yang Xiang","orcid":"https://orcid.org/0000-0002-7120-5842"},"institutions":[{"id":"https://openalex.org/I891191580","display_name":"Aalborg University","ror":"https://ror.org/04m5j1k67","country_code":"DK","type":"education","lineage":["https://openalex.org/I891191580"]}],"countries":["DK"],"is_corresponding":true,"raw_author_name":"Yang Xiang","raw_affiliation_strings":["Aalborg University,Audio Analysis Lab, CREATE,Aalbory,Denmark","Audio Analysis Lab, CREATE, Aalborg University, Aalbory, Denmark","Capturi A/S, Aarhus, Denmark"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Aalborg University,Audio Analysis Lab, CREATE,Aalbory,Denmark","institution_ids":["https://openalex.org/I891191580"]},{"raw_affiliation_string":"Audio Analysis Lab, CREATE, Aalborg University, Aalbory, Denmark","institution_ids":["https://openalex.org/I891191580"]},{"raw_affiliation_string":"Capturi A/S, Aarhus, Denmark","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063946978","display_name":"Jesper Lisby H\u00f8jvang","orcid":"https://orcid.org/0009-0002-6278-9353"},"institutions":[{"id":"https://openalex.org/I4210124916","display_name":"Capnova (Denmark)","ror":"https://ror.org/02zttjj06","country_code":"DK","type":"company","lineage":["https://openalex.org/I4210124916"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Jesper Lisby Hojvang","raw_affiliation_strings":["Capturi A/S,Aarhus,Denmark","Capturi A/S, Aarhus, Denmark"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Capturi A/S,Aarhus,Denmark","institution_ids":["https://openalex.org/I4210124916"]},{"raw_affiliation_string":"Capturi A/S, Aarhus, Denmark","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054964567","display_name":"Morten H\u00f8jfeldt Rasmussen","orcid":null},"institutions":[{"id":"https://openalex.org/I4210124916","display_name":"Capnova (Denmark)","ror":"https://ror.org/02zttjj06","country_code":"DK","type":"company","lineage":["https://openalex.org/I4210124916"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Morten Hojfeldt Rasmussen","raw_affiliation_strings":["Capturi A/S,Aarhus,Denmark","Capturi A/S, Aarhus, Denmark"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Capturi A/S,Aarhus,Denmark","institution_ids":["https://openalex.org/I4210124916"]},{"raw_affiliation_string":"Capturi A/S, Aarhus, Denmark","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5026658144","display_name":"Mads Gr\u00e6sb\u00f8ll Christensen","orcid":"https://orcid.org/0000-0003-3586-7969"},"institutions":[{"id":"https://openalex.org/I891191580","display_name":"Aalborg University","ror":"https://ror.org/04m5j1k67","country_code":"DK","type":"education","lineage":["https://openalex.org/I891191580"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Mads Grasboll Christensen","raw_affiliation_strings":["Aalborg University,Audio Analysis Lab, CREATE,Aalbory,Denmark","Audio Analysis Lab, CREATE, Aalborg University, Aalbory, Denmark"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Aalborg University,Audio Analysis Lab, CREATE,Aalbory,Denmark","institution_ids":["https://openalex.org/I891191580"]},{"raw_affiliation_string":"Audio Analysis Lab, CREATE, Aalborg University, Aalbory, Denmark","institution_ids":["https://openalex.org/I891191580"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5029253113"],"corresponding_institution_ids":["https://openalex.org/I891191580"],"apc_list":null,"apc_paid":null,"fwci":0.9716,"has_fulltext":true,"cited_by_count":9,"citation_normalized_percentile":{"value":0.76266918,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"381","last_page":"385"},"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/T10283","display_name":"Hearing Loss and Rehabilitation","score":0.9968000054359436,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T11233","display_name":"Advanced Adaptive Filtering Techniques","score":0.9916999936103821,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"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/autoencoder","display_name":"Autoencoder","score":0.7774991989135742},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6365139484405518},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5561672449111938},{"id":"https://openalex.org/keywords/non-negative-matrix-factorization","display_name":"Non-negative matrix factorization","score":0.5537815093994141},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.532339870929718},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.5235775709152222},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.5208602547645569},{"id":"https://openalex.org/keywords/latent-variable","display_name":"Latent variable","score":0.4840778410434723},{"id":"https://openalex.org/keywords/speech-enhancement","display_name":"Speech enhancement","score":0.455576092004776},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4275110960006714},{"id":"https://openalex.org/keywords/noise-measurement","display_name":"Noise measurement","score":0.4125884175300598},{"id":"https://openalex.org/keywords/matrix-decomposition","display_name":"Matrix decomposition","score":0.40408390760421753},{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.29771995544433594},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.06356364488601685}],"concepts":[{"id":"https://openalex.org/C101738243","wikidata":"https://www.wikidata.org/wiki/Q786435","display_name":"Autoencoder","level":3,"score":0.7774991989135742},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6365139484405518},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5561672449111938},{"id":"https://openalex.org/C152671427","wikidata":"https://www.wikidata.org/wiki/Q10843505","display_name":"Non-negative matrix factorization","level":4,"score":0.5537815093994141},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.532339870929718},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.5235775709152222},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.5208602547645569},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.4840778410434723},{"id":"https://openalex.org/C2776182073","wikidata":"https://www.wikidata.org/wiki/Q7575395","display_name":"Speech enhancement","level":3,"score":0.455576092004776},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4275110960006714},{"id":"https://openalex.org/C29265498","wikidata":"https://www.wikidata.org/wiki/Q7047719","display_name":"Noise measurement","level":3,"score":0.4125884175300598},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.40408390760421753},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.29771995544433594},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.06356364488601685},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp43922.2022.9747036","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9747036","pdf_url":null,"source":{"id":"https://openalex.org/S4363607702","display_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},{"id":"pmh:oai:pure.atira.dk:publications/5677b3e5-0835-4061-b9b9-75ff9a347ac0","is_oa":true,"landing_page_url":"https://vbn.aau.dk/da/publications/5677b3e5-0835-4061-b9b9-75ff9a347ac0","pdf_url":"http://arxiv.org/pdf/2201.09875.pdf","source":{"id":"https://openalex.org/S4306401731","display_name":"VBN Forskningsportal (Aalborg Universitet)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I891191580","host_organization_name":"Aalborg University","host_organization_lineage":["https://openalex.org/I891191580"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Xiang , Y , H\u00f8jvang , J L , Rasmussen , M H & Christensen , M G 2022 , A Bayesian Permutation Training Deep Representation Learning Method for Speech Enhancement with Variational Autoencoder . in ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE , ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings , pp. 381-385 , 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 , Virtual, Online , Singapore , 23/05/2022 . https://doi.org/10.1109/ICASSP43922.2022.9747036","raw_type":"contributionToPeriodical"}],"best_oa_location":{"id":"pmh:oai:pure.atira.dk:publications/5677b3e5-0835-4061-b9b9-75ff9a347ac0","is_oa":true,"landing_page_url":"https://vbn.aau.dk/da/publications/5677b3e5-0835-4061-b9b9-75ff9a347ac0","pdf_url":"http://arxiv.org/pdf/2201.09875.pdf","source":{"id":"https://openalex.org/S4306401731","display_name":"VBN Forskningsportal (Aalborg Universitet)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I891191580","host_organization_name":"Aalborg University","host_organization_lineage":["https://openalex.org/I891191580"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Xiang , Y , H\u00f8jvang , J L , Rasmussen , M H & Christensen , M G 2022 , A Bayesian Permutation Training Deep Representation Learning Method for Speech Enhancement with Variational Autoencoder . in ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE , ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings , pp. 381-385 , 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 , Virtual, Online , Singapore , 23/05/2022 . https://doi.org/10.1109/ICASSP43922.2022.9747036","raw_type":"contributionToPeriodical"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320334111","display_name":"Innovation Fund","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4225263883.pdf","grobid_xml":"https://content.openalex.org/works/W4225263883.grobid-xml"},"referenced_works_count":34,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W1552314771","https://openalex.org/W1959608418","https://openalex.org/W1974387177","https://openalex.org/W2031647436","https://openalex.org/W2044893557","https://openalex.org/W2051428568","https://openalex.org/W2069681747","https://openalex.org/W2141998673","https://openalex.org/W2159202424","https://openalex.org/W2190061941","https://openalex.org/W2758785877","https://openalex.org/W2766672686","https://openalex.org/W2900450731","https://openalex.org/W2901552243","https://openalex.org/W2952218014","https://openalex.org/W2962850167","https://openalex.org/W2962866211","https://openalex.org/W2963364041","https://openalex.org/W2964058413","https://openalex.org/W3031135612","https://openalex.org/W3096900631","https://openalex.org/W3099330747","https://openalex.org/W3100968126","https://openalex.org/W3130335839","https://openalex.org/W3131332223","https://openalex.org/W3161140524","https://openalex.org/W4232282348","https://openalex.org/W4253928870","https://openalex.org/W6631190155","https://openalex.org/W6640963894","https://openalex.org/W6741347854","https://openalex.org/W6745117592","https://openalex.org/W6785473800"],"related_works":["https://openalex.org/W2127243424","https://openalex.org/W4390394189","https://openalex.org/W2037504162","https://openalex.org/W2539013788","https://openalex.org/W2792706544","https://openalex.org/W1568451138","https://openalex.org/W2156699640","https://openalex.org/W2045265907","https://openalex.org/W2972997031","https://openalex.org/W34555840"],"abstract_inverted_index":{"Recently,":[0],"variational":[1,83],"autoencoder":[2],"(VAE),":[3],"a":[4,81,183],"deep":[5,195],"representation":[6],"learning":[7],"(DRL)":[8],"model,":[9],"has":[10],"been":[11],"used":[12],"to":[13,19,32,119,144,151],"perform":[14],"speech":[15,34,65,100,122,146,171,189],"enhancement":[16],"(SE).":[17],"However,":[18],"the":[20,41,50,64,71,106,112,117,131,138,164,177,193],"best":[21],"of":[22,49,154],"our":[23],"knowledge,":[24],"current":[25],"VAE-based":[26,60,133],"SE":[27,134,199],"methods":[28,61],"only":[29,169],"apply":[30,116],"VAE":[31,91,118],"model":[33,120],"signal,":[35,179],"while":[36],"noise":[37,67,102,124,148,173],"is":[38,57,127],"modeled":[39],"using":[40,55],"traditional":[42],"non-negative":[43],"matrix":[44],"factorization":[45],"(NMF)":[46],"model.":[47],"One":[48],"most":[51],"important":[52],"reasons":[53],"for":[54,86,157],"NMF":[56],"that":[58,90,111,163],"these":[59],"cannot":[62],"disentangle":[63,99,170],"and":[66,97,101,123,147,172,188],"latent":[68,103,155,174],"variables":[69,104,156,175],"from":[70,105,130,176],"observed":[72,107,178],"signal.":[73,108],"Based":[74],"on":[75],"Bayesian":[76],"theory,":[77],"this":[78],"paper":[79],"derives":[80],"novel":[82],"lower":[84],"bound":[85],"VAE,":[87],"which":[88,126],"ensures":[89],"can":[92,98,115,142,167],"be":[93],"trained":[94],"in":[95],"supervision,":[96],"This":[109],"means":[110],"proposed":[113,139,165],"method":[114,141,166],"both":[121],"signals,":[125],"totally":[128],"different":[129,152],"previous":[132],"works.":[135],"More":[136],"specifically,":[137],"DRL":[140],"learn":[143],"impose":[145],"signal":[149],"priors":[150],"sets":[153],"SE.":[158],"The":[159],"experimental":[160],"results":[161],"show":[162],"not":[168],"but":[180],"also":[181],"obtain":[182],"higher":[184],"scale-invariant":[185],"signal-to-distortion":[186],"ratio":[187],"quality":[190],"score":[191],"than":[192],"similar":[194],"neural":[196],"network-based":[197],"(DNN)":[198],"method.":[200]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":2}],"updated_date":"2026-04-29T09:16:38.111599","created_date":"2025-10-10T00:00:00"}