{"id":"https://openalex.org/W2996743179","doi":"https://doi.org/10.1109/tsp.2021.3066038","title":"Mixture of Inference Networks for VAE-Based Audio-Visual Speech Enhancement","display_name":"Mixture of Inference Networks for VAE-Based Audio-Visual Speech Enhancement","publication_year":2021,"publication_date":"2021-01-01","ids":{"openalex":"https://openalex.org/W2996743179","doi":"https://doi.org/10.1109/tsp.2021.3066038","mag":"2996743179"},"language":"en","primary_location":{"id":"doi:10.1109/tsp.2021.3066038","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tsp.2021.3066038","pdf_url":null,"source":{"id":"https://openalex.org/S168680287","display_name":"IEEE Transactions on Signal Processing","issn_l":"1053-587X","issn":["1053-587X","1941-0476"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Signal Processing","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/1912.10647","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5062612015","display_name":"Mostafa Sadeghi","orcid":"https://orcid.org/0000-0002-0272-8017"},"institutions":[{"id":"https://openalex.org/I4210127166","display_name":"Centre Inria de l'Universit\u00e9 de Lorraine","ror":"https://ror.org/03fcjvn64","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1326498283","https://openalex.org/I4210127166"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Mostafa Sadeghi","raw_affiliation_strings":["Multispeech Team, Inria Research Centre Nancy Grand Est, Villers-les-Nancy, France","[Multispeech Team, Inria Research Centre Nancy Grand Est, Villers-les-Nancy, France]"],"raw_orcid":"https://orcid.org/0000-0002-0272-8017","affiliations":[{"raw_affiliation_string":"Multispeech Team, Inria Research Centre Nancy Grand Est, Villers-les-Nancy, France","institution_ids":["https://openalex.org/I4210127166"]},{"raw_affiliation_string":"[Multispeech Team, Inria Research Centre Nancy Grand Est, Villers-les-Nancy, France]","institution_ids":["https://openalex.org/I4210127166"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066621495","display_name":"Xavier Alameda-Pineda","orcid":"https://orcid.org/0000-0002-5354-1084"},"institutions":[{"id":"https://openalex.org/I4210101348","display_name":"Centre Inria de l'Universit\u00e9 Grenoble Alpes","ror":"https://ror.org/00n8d6z93","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1326498283","https://openalex.org/I4210101348"]},{"id":"https://openalex.org/I899635006","display_name":"Universit\u00e9 Grenoble Alpes","ror":"https://ror.org/02rx3b187","country_code":"FR","type":"education","lineage":["https://openalex.org/I899635006"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Xavier Alameda-Pineda","raw_affiliation_strings":["Perception Team, Inria Centre de Recherche Grenoble Rhone-Alpes, Montbonnot, France"],"raw_orcid":"https://orcid.org/0000-0002-5354-1084","affiliations":[{"raw_affiliation_string":"Perception Team, Inria Centre de Recherche Grenoble Rhone-Alpes, Montbonnot, France","institution_ids":["https://openalex.org/I4210101348","https://openalex.org/I899635006"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.5579,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.62818281,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":95},"biblio":{"volume":"69","issue":null,"first_page":"1899","last_page":"1909"},"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/T11233","display_name":"Advanced Adaptive Filtering Techniques","score":0.9923999905586243,"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"}},{"id":"https://openalex.org/T10283","display_name":"Hearing Loss and Rehabilitation","score":0.9904000163078308,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7844594717025757},{"id":"https://openalex.org/keywords/initialization","display_name":"Initialization","score":0.6754050850868225},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6686305403709412},{"id":"https://openalex.org/keywords/latent-variable","display_name":"Latent variable","score":0.6149383783340454},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5897142887115479},{"id":"https://openalex.org/keywords/mixture-model","display_name":"Mixture model","score":0.586554765701294},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5403587818145752},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5125069618225098},{"id":"https://openalex.org/keywords/autoencoder","display_name":"Autoencoder","score":0.5100165009498596},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.5025320053100586},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.4266631007194519},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4129515588283539},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.34106385707855225},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.33313173055648804},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.2777174115180969},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.12549152970314026}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7844594717025757},{"id":"https://openalex.org/C114466953","wikidata":"https://www.wikidata.org/wiki/Q6034165","display_name":"Initialization","level":2,"score":0.6754050850868225},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6686305403709412},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.6149383783340454},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5897142887115479},{"id":"https://openalex.org/C61224824","wikidata":"https://www.wikidata.org/wiki/Q2260434","display_name":"Mixture model","level":2,"score":0.586554765701294},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5403587818145752},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5125069618225098},{"id":"https://openalex.org/C101738243","wikidata":"https://www.wikidata.org/wiki/Q786435","display_name":"Autoencoder","level":3,"score":0.5100165009498596},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.5025320053100586},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.4266631007194519},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4129515588283539},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.34106385707855225},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33313173055648804},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.2777174115180969},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.12549152970314026},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":6,"locations":[{"id":"doi:10.1109/tsp.2021.3066038","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tsp.2021.3066038","pdf_url":null,"source":{"id":"https://openalex.org/S168680287","display_name":"IEEE Transactions on Signal Processing","issn_l":"1053-587X","issn":["1053-587X","1941-0476"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Signal Processing","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:1912.10647","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1912.10647","pdf_url":"https://arxiv.org/pdf/1912.10647","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"mag:2996743179","is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/1912.10647","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"pmh:oai:HAL:hal-02926172v1","is_oa":true,"landing_page_url":"https://hal.inria.fr/hal-02926172","pdf_url":null,"source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"ISSN: 1053-587X","raw_type":"info:eu-repo/semantics/article"},{"id":"pmh:oai:HAL:hal-02926172v2","is_oa":true,"landing_page_url":"https://inria.hal.science/hal-02926172","pdf_url":null,"source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"IEEE Transactions on Signal Processing, 2021, 69, pp.1899-1909. ⟨10.1109/TSP.2021.3066038⟩","raw_type":"info:eu-repo/semantics/article"},{"id":"doi:10.48550/arxiv.1912.10647","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1912.10647","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:1912.10647","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1912.10647","pdf_url":"https://arxiv.org/pdf/1912.10647","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"score":0.75,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[{"id":"https://openalex.org/G1041279674","display_name":"MIAI @ Grenoble Alpes","funder_award_id":"ANR-19-P3IA-0003","funder_id":"https://openalex.org/F4320320883","funder_display_name":"Agence Nationale de la Recherche"},{"id":"https://openalex.org/G5495492380","display_name":"Multi-Modal Multi-person Low-Level Learning for Robot Interaction","funder_award_id":"ANR-19-CE33-0008","funder_id":"https://openalex.org/F4320320883","funder_display_name":"Agence Nationale de la Recherche"}],"funders":[{"id":"https://openalex.org/F4320320883","display_name":"Agence Nationale de la Recherche","ror":"https://ror.org/00rbzpz17"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":50,"referenced_works":["https://openalex.org/W1495679096","https://openalex.org/W1504438288","https://openalex.org/W1552314771","https://openalex.org/W1663973292","https://openalex.org/W1959608418","https://openalex.org/W1968939597","https://openalex.org/W2013608223","https://openalex.org/W2038010270","https://openalex.org/W2039844283","https://openalex.org/W2044893557","https://openalex.org/W2054852564","https://openalex.org/W2096698681","https://openalex.org/W2115109087","https://openalex.org/W2127851351","https://openalex.org/W2128653836","https://openalex.org/W2141998673","https://openalex.org/W2158291955","https://openalex.org/W2591920181","https://openalex.org/W2675830001","https://openalex.org/W2747876639","https://openalex.org/W2766672686","https://openalex.org/W2911579794","https://openalex.org/W2918296821","https://openalex.org/W2922004249","https://openalex.org/W2943237054","https://openalex.org/W2962866211","https://openalex.org/W2962897886","https://openalex.org/W2963082324","https://openalex.org/W2963375116","https://openalex.org/W2964121744","https://openalex.org/W2964171275","https://openalex.org/W2965390715","https://openalex.org/W2975955714","https://openalex.org/W2989954484","https://openalex.org/W3015791137","https://openalex.org/W3049360585","https://openalex.org/W3082851515","https://openalex.org/W3100968126","https://openalex.org/W3127686677","https://openalex.org/W3213412873","https://openalex.org/W4245919820","https://openalex.org/W6631190155","https://openalex.org/W6639732818","https://openalex.org/W6640963894","https://openalex.org/W6683202405","https://openalex.org/W6743400243","https://openalex.org/W6783123723","https://openalex.org/W6785473800","https://openalex.org/W6789826613","https://openalex.org/W6790096238"],"related_works":["https://openalex.org/W3137758952","https://openalex.org/W3049360585","https://openalex.org/W2982264772","https://openalex.org/W2102863932","https://openalex.org/W3160567113","https://openalex.org/W3125709657","https://openalex.org/W2165759752","https://openalex.org/W3100968126","https://openalex.org/W2965390715","https://openalex.org/W2964121744","https://openalex.org/W2141998673","https://openalex.org/W1552314771","https://openalex.org/W2131459862","https://openalex.org/W2403836935","https://openalex.org/W2156043751","https://openalex.org/W2022798421","https://openalex.org/W2090994663","https://openalex.org/W2347453903","https://openalex.org/W1546363070","https://openalex.org/W2936295285"],"abstract_inverted_index":{"We":[0],"address":[1],"unsupervised":[2,178],"audio-visual":[3,95,244],"speech":[4,17,59],"enhancement":[5,26],"based":[6],"on":[7],"variational":[8,127,212],"autoencoders":[9],"(VAEs),":[10],"where":[11],"the":[12,29,49,53,57,61,66,78,81,85,105,113,122,140,143,168,172,186,193,200,219,224,229,232,238],"prior":[13],"distribution":[14],"of":[15,48,56,80,124,142,151],"clean":[16,58,89],"spectrogram":[18],"is":[19,34,63,70,73,146,162,177,207,215],"simulated":[20],"using":[21,77],"an":[22,101],"encoder-decoder":[23],"architecture.":[24],"At":[25],"(test)":[27],"time,":[28,199],"trained":[30,82],"generative":[31,54],"model":[32,39],"(decoder)":[33],"combined":[35],"with":[36],"a":[37,149,183],"noise":[38],"whose":[40],"parameters":[41],"need":[42],"to":[43,190,217,223],"be":[44],"estimated.":[45],"The":[46,159],"initialization":[47,103],"latent":[50,144],"variables":[51,145],"describing":[52],"process":[55],"via":[60],"decoder,":[62,185],"crucial,":[64],"as":[65,92,148,179,241,243],"overall":[67,187],"inference":[68,125,213,226],"problem":[69],"non-convex.":[71],"This":[72],"usually":[74],"done":[75],"by":[76],"output":[79,155],"encoder":[83,131],"given":[84],"noisy":[86],"audio":[87,135,173],"and":[88,136,139,165,174,228],"visual":[90,137,175,201,205],"data":[91],"input.":[93],"Current":[94],"VAE":[96],"models":[97],"do":[98],"not":[99],"provide":[100],"effective":[102],"because":[104],"two":[106,152,194],"modalities":[107],"are":[108],"tightly":[109],"coupled":[110],"(concatenated)":[111],"in":[112],"associated":[114],"architectures.":[115],"To":[116],"overcome":[117],"this":[118],"issue,":[119],"we":[120],"introduce":[121],"mixture":[123,150,160],"networks":[126,132],"autoencoder":[128],"(MIN-VAE).":[129],"Two":[130],"input,":[133],"respectively,":[134],"data,":[138,206],"posterior":[141],"modeled":[147],"Gaussian":[153],"distributions":[154],"from":[156],"each":[157],"encoder.":[158],"variable":[161],"also":[163],"latent,":[164],"therefore":[166],"learning":[167],"optimal":[169],"balance":[170],"between":[171],"encoders":[176],"well.":[180],"By":[181],"training":[182],"shared":[184],"network":[188],"learns":[189],"adaptively":[191],"fuse":[192],"modalities.":[195],"Moreover,":[196],"at":[197],"test":[198],"encoder,":[202],"taking":[203],"(clean)":[204],"used":[208],"for":[209],"initialization.":[210],"A":[211],"approach":[214],"derived":[216],"train":[218],"proposed":[220],"model.":[221],"Thanks":[222],"novel":[225],"procedure":[227],"robust":[230],"initialization,":[231],"MIN-VAE":[233],"exhibits":[234],"superior":[235],"performance":[236],"than":[237],"standard":[239],"audio-only":[240],"well":[242],"counterparts.":[245]},"counts_by_year":[{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":2}],"updated_date":"2026-07-15T18:14:33.161393","created_date":"2022-07-26T00:00:00"}