{"id":"https://openalex.org/W4406860666","doi":"https://doi.org/10.1109/apsipaasc63619.2025.10849007","title":"Bluemarble: Bridging Latent Uncertainty in Articulatory-to-Speech Synthesis with a Learned Codebook","display_name":"Bluemarble: Bridging Latent Uncertainty in Articulatory-to-Speech Synthesis with a Learned Codebook","publication_year":2024,"publication_date":"2024-12-03","ids":{"openalex":"https://openalex.org/W4406860666","doi":"https://doi.org/10.1109/apsipaasc63619.2025.10849007"},"language":"en","primary_location":{"id":"doi:10.1109/apsipaasc63619.2025.10849007","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc63619.2025.10849007","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5074612173","display_name":"Seyun Um","orcid":"https://orcid.org/0000-0002-2229-6741"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seyun Um","raw_affiliation_strings":["Yonsei University,Department of Electrical and Electronic Engineering,Seoul,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yonsei University,Department of Electrical and Electronic Engineering,Seoul,Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066878875","display_name":"Miseul Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Miseul Kim","raw_affiliation_strings":["Yonsei University,Department of Electrical and Electronic Engineering,Seoul,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yonsei University,Department of Electrical and Electronic Engineering,Seoul,Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087299274","display_name":"Doyeon Kim","orcid":"https://orcid.org/0009-0008-1055-0563"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Doyeon Kim","raw_affiliation_strings":["Yonsei University,Department of Electrical and Electronic Engineering,Seoul,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yonsei University,Department of Electrical and Electronic Engineering,Seoul,Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101203935","display_name":"Hong-Goo Kang","orcid":null},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hong-Goo Kang","raw_affiliation_strings":["Yonsei University,Department of Electrical and Electronic Engineering,Seoul,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yonsei University,Department of Electrical and Electronic Engineering,Seoul,Korea","institution_ids":["https://openalex.org/I193775966"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I193775966"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.29423643,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.9897000193595886,"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/T12031","display_name":"Speech and dialogue systems","score":0.9897000193595886,"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/T10403","display_name":"Phonetics and Phonology Research","score":0.9754999876022339,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9179999828338623,"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/codebook","display_name":"Codebook","score":0.902185320854187},{"id":"https://openalex.org/keywords/bridging","display_name":"Bridging (networking)","score":0.8055214881896973},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6855246424674988},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5537020564079285},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.47245433926582336},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4713147282600403},{"id":"https://openalex.org/keywords/speech-synthesis","display_name":"Speech synthesis","score":0.4215569496154785}],"concepts":[{"id":"https://openalex.org/C127759330","wikidata":"https://www.wikidata.org/wiki/Q637416","display_name":"Codebook","level":2,"score":0.902185320854187},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.8055214881896973},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6855246424674988},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5537020564079285},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.47245433926582336},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4713147282600403},{"id":"https://openalex.org/C14999030","wikidata":"https://www.wikidata.org/wiki/Q16346","display_name":"Speech synthesis","level":2,"score":0.4215569496154785},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/apsipaasc63619.2025.10849007","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc63619.2025.10849007","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","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":33,"referenced_works":["https://openalex.org/W2023407680","https://openalex.org/W2147122824","https://openalex.org/W2255466643","https://openalex.org/W2593414223","https://openalex.org/W2888917702","https://openalex.org/W2952609044","https://openalex.org/W2998572311","https://openalex.org/W3096831136","https://openalex.org/W3097206152","https://openalex.org/W3097945073","https://openalex.org/W3133449202","https://openalex.org/W3175752069","https://openalex.org/W3196475561","https://openalex.org/W3209984917","https://openalex.org/W4205787394","https://openalex.org/W4297841848","https://openalex.org/W4312097445","https://openalex.org/W4372264243","https://openalex.org/W4372341338","https://openalex.org/W4381786045","https://openalex.org/W4385245566","https://openalex.org/W4386132131","https://openalex.org/W4388674325","https://openalex.org/W4390075359","https://openalex.org/W4393969253","https://openalex.org/W6677326919","https://openalex.org/W6677477928","https://openalex.org/W6691096134","https://openalex.org/W6711753470","https://openalex.org/W6783867762","https://openalex.org/W6790742069","https://openalex.org/W6802805937","https://openalex.org/W6802838302"],"related_works":["https://openalex.org/W2293149949","https://openalex.org/W2026099691","https://openalex.org/W4284672201","https://openalex.org/W2377486419","https://openalex.org/W2943202426","https://openalex.org/W2137816434","https://openalex.org/W2163679795","https://openalex.org/W2736714427","https://openalex.org/W2017956276","https://openalex.org/W2136885550"],"abstract_inverted_index":{"In":[0],"this":[1,72],"letter,":[2],"we":[3,74],"present":[4],"Bluemarble,":[5],"a":[6,25,52,76,81,93,97,107,143],"novel":[7],"neural":[8,144],"articulation-to-speech":[9],"(ATS)":[10],"system":[11],"designed":[12],"to":[13,65,85,146],"generate":[14],"high-quality":[15],"speech":[16,32,98,123,149],"from":[17,115],"articulatory":[18,30,63],"recordings.":[19],"Traditional":[20],"ATS":[21],"methods":[22,160],"often":[23],"demand":[24],"substantial":[26],"dataset":[27],"of":[28,48,96],"corresponding":[29],"and":[31,102,120,164],"signal":[33],"pairs":[34],"for":[35,59],"effective":[36],"model":[37,105],"training.":[38],"This":[39],"requirement":[40],"arises":[41],"because":[42],"these":[43],"approaches":[44],"treat":[45],"the":[46,66,87,111,126,148],"estimation":[47],"acoustic":[49],"features":[50,64],"as":[51],"regression":[53],"task,":[54],"which":[55,138],"can":[56],"pose":[57],"difficulties":[58],"accurately":[60],"mapping":[61,108],"latent":[62,89,112],"desired":[67],"speech-related":[68,135],"features.":[69],"To":[70],"tackle":[71],"challenge,":[73],"propose":[75],"method":[77],"that":[78,154],"involves":[79],"employing":[80],"finite":[82],"learned":[83],"codebook":[84],"constrain":[86],"uncertainty":[88],"space.":[90],"Operating":[91],"within":[92],"framework":[94],"consisting":[95],"mel-spectrogram":[99],"encoder,":[100],"quantizer,":[101],"decoder,":[103],"our":[104,155],"establishes":[106],"rule":[109],"between":[110],"embeddings":[113,130],"derived":[114],"electromagnetic":[116],"articulography":[117],"(EMA)":[118],"signals":[119],"quantized":[121],"discrete":[122,136],"tokens.":[124],"During":[125],"inference":[127],"stage,":[128],"EMA":[129],"are":[131,139,169],"first":[132],"transformed":[133],"into":[134,142],"tokens,":[137],"then":[140],"input":[141],"vocoder":[145],"synthesize":[147],"waveform.":[150],"Experimental":[151],"results":[152],"show":[153],"approach":[156],"outperforms":[157],"existing":[158],"state-of-the-art":[159],"in":[161],"both":[162],"qualitative":[163],"quantitative":[165],"assessments.":[166],"Audio":[167],"samples":[168],"available":[170],"online":[171],".1":[172]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}