{"id":"https://openalex.org/W4411724981","doi":"https://doi.org/10.1109/lsp.2025.3583967","title":"Posterior Transition Modeling for Unsupervised Diffusion-Based Speech Enhancement","display_name":"Posterior Transition Modeling for Unsupervised Diffusion-Based Speech Enhancement","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4411724981","doi":"https://doi.org/10.1109/lsp.2025.3583967"},"language":"en","primary_location":{"id":"doi:10.1109/lsp.2025.3583967","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2025.3583967","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"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 Signal Processing Letters","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://hal.science/hal-05135495/document","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/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"government","lineage":["https://openalex.org/I1294671590"]},{"id":"https://openalex.org/I4210121838","display_name":"Laboratoire Lorrain de Recherche en Informatique et ses Applications","ror":"https://ror.org/02vnf0c38","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I277688954","https://openalex.org/I4210107720","https://openalex.org/I4210121838","https://openalex.org/I4210159245","https://openalex.org/I90183372"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Mostafa Sadeghi","raw_affiliation_strings":["Multispeech Team, Université de Lorraine, CNRS, Inria, Loria, Nancy, France"],"raw_orcid":"https://orcid.org/0000-0002-0272-8017","affiliations":[{"raw_affiliation_string":"Multispeech Team, Université de Lorraine, CNRS, Inria, Loria, Nancy, France","institution_ids":["https://openalex.org/I1294671590","https://openalex.org/I4210121838"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092912265","display_name":"Jean-Eudes Ayilo","orcid":null},"institutions":[{"id":"https://openalex.org/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"government","lineage":["https://openalex.org/I1294671590"]},{"id":"https://openalex.org/I4210121838","display_name":"Laboratoire Lorrain de Recherche en Informatique et ses Applications","ror":"https://ror.org/02vnf0c38","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I277688954","https://openalex.org/I4210107720","https://openalex.org/I4210121838","https://openalex.org/I4210159245","https://openalex.org/I90183372"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Jean-Eudes Ayilo","raw_affiliation_strings":["Multispeech Team, Université de Lorraine, CNRS, Inria, Loria, Nancy, France"],"raw_orcid":"https://orcid.org/0009-0005-4017-8078","affiliations":[{"raw_affiliation_string":"Multispeech Team, Université de Lorraine, CNRS, Inria, Loria, Nancy, France","institution_ids":["https://openalex.org/I1294671590","https://openalex.org/I4210121838"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064988262","display_name":"Romain Serizel","orcid":"https://orcid.org/0000-0002-6848-0114"},"institutions":[{"id":"https://openalex.org/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"government","lineage":["https://openalex.org/I1294671590"]},{"id":"https://openalex.org/I4210121838","display_name":"Laboratoire Lorrain de Recherche en Informatique et ses Applications","ror":"https://ror.org/02vnf0c38","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I277688954","https://openalex.org/I4210107720","https://openalex.org/I4210121838","https://openalex.org/I4210159245","https://openalex.org/I90183372"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Romain Serizel","raw_affiliation_strings":["Multispeech Team, Université de Lorraine, CNRS, Inria, Loria, Nancy, France"],"raw_orcid":"https://orcid.org/0000-0002-6848-0114","affiliations":[{"raw_affiliation_string":"Multispeech Team, Université de Lorraine, CNRS, Inria, Loria, Nancy, France","institution_ids":["https://openalex.org/I1294671590","https://openalex.org/I4210121838"]}]},{"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/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"government","lineage":["https://openalex.org/I1294671590"]},{"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/I4210149092","display_name":"Laboratoire Jean Kuntzmann","ror":"https://ror.org/04ett5b41","country_code":"FR","type":"facility","lineage":["https://openalex.org/I106785703","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I4210149092","https://openalex.org/I899635006","https://openalex.org/I899635006"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Xavier Alameda-Pineda","raw_affiliation_strings":["RobotLearn Team, Inria, Université Grenoble Alpes, CNRS, LJK, Montbonnot, France"],"raw_orcid":"https://orcid.org/0000-0002-5354-1084","affiliations":[{"raw_affiliation_string":"RobotLearn Team, Inria, Université Grenoble Alpes, CNRS, LJK, Montbonnot, France","institution_ids":["https://openalex.org/I1294671590","https://openalex.org/I4210101348","https://openalex.org/I4210149092"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.2879,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.79939468,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":97},"biblio":{"volume":"32","issue":null,"first_page":"2694","last_page":"2698"},"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":0.9973000288009644,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9965000152587891,"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":0.5732746124267578},{"id":"https://openalex.org/keywords/speech-enhancement","display_name":"Speech enhancement","score":0.5501474142074585},{"id":"https://openalex.org/keywords/transition","display_name":"Transition (genetics)","score":0.5221555233001709},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5044714212417603},{"id":"https://openalex.org/keywords/diffusion","display_name":"Diffusion","score":0.48455682396888733},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4135409891605377},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3877057433128357},{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.1620216965675354},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.13484877347946167},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.09074351191520691}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5732746124267578},{"id":"https://openalex.org/C2776182073","wikidata":"https://www.wikidata.org/wiki/Q7575395","display_name":"Speech enhancement","level":3,"score":0.5501474142074585},{"id":"https://openalex.org/C194232998","wikidata":"https://www.wikidata.org/wiki/Q1606712","display_name":"Transition (genetics)","level":3,"score":0.5221555233001709},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5044714212417603},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.48455682396888733},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4135409891605377},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3877057433128357},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.1620216965675354},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.13484877347946167},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.09074351191520691},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C97355855","wikidata":"https://www.wikidata.org/wiki/Q11473","display_name":"Thermodynamics","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/lsp.2025.3583967","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2025.3583967","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"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 Signal Processing Letters","raw_type":"journal-article"},{"id":"pmh:oai:HAL:hal-05135495v1","is_oa":true,"landing_page_url":"https://hal.science/hal-05135495","pdf_url":"https://hal.science/hal-05135495/document","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"IEEE Signal Processing Letters, 2025, 32, pp.2694-2698. ⟨10.1109/LSP.2025.3583967⟩","raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"pmh:oai:HAL:hal-05135495v1","is_oa":true,"landing_page_url":"https://hal.science/hal-05135495","pdf_url":"https://hal.science/hal-05135495/document","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"IEEE Signal Processing Letters, 2025, 32, pp.2694-2698. ⟨10.1109/LSP.2025.3583967⟩","raw_type":"info:eu-repo/semantics/article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1102524033","display_name":"Robust and Efficient Deep Learning based Audiovisual Speech Enhancement","funder_award_id":"ANR-22-CE23-0026","funder_id":"https://openalex.org/F4320320883","funder_display_name":"Agence Nationale de la Recherche"},{"id":"https://openalex.org/G3780308692","display_name":null,"funder_award_id":"ANR-22-CE23-0026-01","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":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4411724981.pdf","grobid_xml":"https://content.openalex.org/works/W4411724981.grobid-xml"},"referenced_works_count":34,"referenced_works":["https://openalex.org/W891534129","https://openalex.org/W1552314771","https://openalex.org/W1991111872","https://openalex.org/W2015810892","https://openalex.org/W2044893557","https://openalex.org/W2516001803","https://openalex.org/W2603567530","https://openalex.org/W2952218014","https://openalex.org/W2962866211","https://openalex.org/W2964058413","https://openalex.org/W3005621653","https://openalex.org/W3099330747","https://openalex.org/W4212863985","https://openalex.org/W4221144097","https://openalex.org/W4232282348","https://openalex.org/W4281820413","https://openalex.org/W4360884927","https://openalex.org/W4375869120","https://openalex.org/W4380434618","https://openalex.org/W4384080510","https://openalex.org/W4385822783","https://openalex.org/W4392903364","https://openalex.org/W4392910516","https://openalex.org/W4403116286","https://openalex.org/W4403126355","https://openalex.org/W4403818631","https://openalex.org/W4405974256","https://openalex.org/W6786375611","https://openalex.org/W6803078664","https://openalex.org/W6844364769","https://openalex.org/W6847029043","https://openalex.org/W6851069163","https://openalex.org/W6873337241","https://openalex.org/W6908971994"],"related_works":["https://openalex.org/W1630865680","https://openalex.org/W4214692512","https://openalex.org/W2373767407","https://openalex.org/W2516933085","https://openalex.org/W3168109306","https://openalex.org/W1727353606","https://openalex.org/W2292168444","https://openalex.org/W2360747494","https://openalex.org/W3096184950","https://openalex.org/W4231424160"],"abstract_inverted_index":{"We":[0],"explore":[1],"unsupervised":[2,121],"speech":[3,24,89],"enhancement":[4,108],"using":[5,22],"diffusion":[6,20,57,64,84],"models":[7],"as":[8],"expressive":[9],"generative":[10],"priors":[11],"for":[12,77],"clean":[13],"speech.":[14],"Existing":[15],"approaches":[16],"guide":[17],"the":[18,33,51,63,67,75,87,101],"reverse":[19,53],"process":[21,85],"noisy":[23,88],"through":[25],"an":[26],"approximate,":[27],"noise-perturbed":[28],"likelihood":[29,97],"score,":[30],"combined":[31],"with":[32,66],"unconditional":[34],"score":[35],"via":[36],"a":[37,71,83,92],"trade-off":[38],"hyperparameter.":[39],"In":[40],"this":[41],"work,":[42],"we":[43],"propose":[44],"two":[45],"alternative":[46],"algorithms":[47],"that":[48],"directly":[49],"model":[50,69],"conditional":[52],"transition":[54],"distribution":[55],"of":[56],"states.":[58],"The":[59,80],"first":[60],"method":[61],"integrates":[62],"prior":[65],"observation":[68],"in":[70],"principled":[72],"way,":[73],"removing":[74],"need":[76],"hyperparameter":[78],"tuning.":[79],"second":[81],"defines":[82],"over":[86],"itself,":[90],"yielding":[91],"fully":[93],"tractable":[94],"and":[95,103,110,120],"exact":[96],"score.":[98],"Experiments":[99],"on":[100],"WSJ0-QUT":[102],"VoiceBank-DEMAND":[104],"datasets":[105],"demonstrate":[106],"improved":[107],"metrics":[109],"greater":[111],"robustness":[112],"to":[113,117],"domain":[114],"shifts":[115],"compared":[116],"both":[118],"supervised":[119],"baselines.":[122]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2025-10-10T00:00:00"}