{"id":"https://openalex.org/W2963581463","doi":"https://doi.org/10.1109/icassp.2019.8683480","title":"End-to-end Feedback Loss in Speech Chain Framework via Straight-through Estimator","display_name":"End-to-end Feedback Loss in Speech Chain Framework via Straight-through Estimator","publication_year":2019,"publication_date":"2019-04-17","ids":{"openalex":"https://openalex.org/W2963581463","doi":"https://doi.org/10.1109/icassp.2019.8683480","mag":"2963581463"},"language":"en","primary_location":{"id":"doi:10.1109/icassp.2019.8683480","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2019.8683480","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"preprint","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/A5038296765","display_name":"Andros Tjandra","orcid":"https://orcid.org/0000-0003-1246-5908"},"institutions":[{"id":"https://openalex.org/I75917431","display_name":"Nara Institute of Science and Technology","ror":"https://ror.org/05bhada84","country_code":"JP","type":"education","lineage":["https://openalex.org/I75917431"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Andros Tjandra","raw_affiliation_strings":["Nara Institute of Science and Technology, Japan"],"affiliations":[{"raw_affiliation_string":"Nara Institute of Science and Technology, Japan","institution_ids":["https://openalex.org/I75917431"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040108974","display_name":"Sakriani Sakti","orcid":"https://orcid.org/0000-0001-5509-8963"},"institutions":[{"id":"https://openalex.org/I75917431","display_name":"Nara Institute of Science and Technology","ror":"https://ror.org/05bhada84","country_code":"JP","type":"education","lineage":["https://openalex.org/I75917431"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Sakriani Sakti","raw_affiliation_strings":["Nara Institute of Science and Technology, Japan"],"affiliations":[{"raw_affiliation_string":"Nara Institute of Science and Technology, Japan","institution_ids":["https://openalex.org/I75917431"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5020994673","display_name":"Satoshi Nakamura","orcid":"https://orcid.org/0000-0001-6956-3803"},"institutions":[{"id":"https://openalex.org/I75917431","display_name":"Nara Institute of Science and Technology","ror":"https://ror.org/05bhada84","country_code":"JP","type":"education","lineage":["https://openalex.org/I75917431"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Satoshi Nakamura","raw_affiliation_strings":["Nara Institute of Science and Technology, Japan"],"affiliations":[{"raw_affiliation_string":"Nara Institute of Science and Technology, Japan","institution_ids":["https://openalex.org/I75917431"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5038296765"],"corresponding_institution_ids":["https://openalex.org/I75917431"],"apc_list":null,"apc_paid":null,"fwci":4.30129687,"has_fulltext":false,"cited_by_count":31,"citation_normalized_percentile":{"value":0.9495854,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"6281","last_page":"6285"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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":0.9998999834060669,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9994000196456909,"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/T10028","display_name":"Topic Modeling","score":0.9987000226974487,"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":0.7744736075401306},{"id":"https://openalex.org/keywords/transcription","display_name":"Transcription (linguistics)","score":0.7185947299003601},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6208270788192749},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.5657833218574524},{"id":"https://openalex.org/keywords/estimator","display_name":"Estimator","score":0.5170162916183472},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.41492778062820435},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.10295528173446655}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7744736075401306},{"id":"https://openalex.org/C179926584","wikidata":"https://www.wikidata.org/wiki/Q207714","display_name":"Transcription (linguistics)","level":2,"score":0.7185947299003601},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6208270788192749},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.5657833218574524},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.5170162916183472},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41492778062820435},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.10295528173446655},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp.2019.8683480","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2019.8683480","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions","score":0.47999998927116394}],"awards":[],"funders":[{"id":"https://openalex.org/F4320334764","display_name":"Japan Society for the Promotion of Science","ror":"https://ror.org/00hhkn466"},{"id":"https://openalex.org/F4320338335","display_name":"H2020 European Research Council","ror":"https://ror.org/0472cxd90"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":31,"referenced_works":["https://openalex.org/W950853366","https://openalex.org/W1524333225","https://openalex.org/W1710082047","https://openalex.org/W1821462560","https://openalex.org/W1902237438","https://openalex.org/W2024490156","https://openalex.org/W2120847449","https://openalex.org/W2133564696","https://openalex.org/W2242818861","https://openalex.org/W2327501763","https://openalex.org/W2526425061","https://openalex.org/W2546938941","https://openalex.org/W2547875792","https://openalex.org/W2548228487","https://openalex.org/W2604184139","https://openalex.org/W2884852625","https://openalex.org/W2952165242","https://openalex.org/W2962699523","https://openalex.org/W2962759037","https://openalex.org/W2962826786","https://openalex.org/W2963216553","https://openalex.org/W2963796886","https://openalex.org/W2964308564","https://openalex.org/W6631362777","https://openalex.org/W6638523607","https://openalex.org/W6679434410","https://openalex.org/W6690026940","https://openalex.org/W6729383884","https://openalex.org/W6730091202","https://openalex.org/W6736356763","https://openalex.org/W6745177358"],"related_works":["https://openalex.org/W3179968364","https://openalex.org/W4287880334","https://openalex.org/W4366700029","https://openalex.org/W2151749779","https://openalex.org/W1999612375","https://openalex.org/W4285230481","https://openalex.org/W2938107654","https://openalex.org/W4385769873","https://openalex.org/W2028097510","https://openalex.org/W2505877856"],"abstract_inverted_index":{"The":[0],"speech":[1,6,27,69,80,93,156],"chain":[2,28,157],"mechanism":[3,29],"integrates":[4],"automatic":[5],"recognition":[7],"(ASR)":[8],"and":[9,40,51,58,82,140,149,182],"text-to-speech":[10],"synthesis":[11],"(TTS)":[12],"modules":[13],"into":[14],"a":[15,26,31,155,163],"single":[16],"cycle":[17],"during":[18],"training.":[19],"In":[20,142],"our":[21,102],"previous":[22,97],"work,":[23],"we":[24,66,99,145,175],"applied":[25],"as":[30],"semi-supervised":[32],"learning.":[33],"It":[34],"provides":[35],"the":[36,55,60,74,79,86,91,105,110,118,121,127,130,138,184,194],"ability":[37],"for":[38,159],"ASR":[39,72,122,131,180,185],"TTS":[41,84,111,139],"to":[42,89,104,126,152,178,193],"assist":[43],"each":[44],"other":[45],"when":[46],"they":[47],"receive":[48],"unpaired":[49],"data":[50],"let":[52],"them":[53],"infer":[54],"missing":[56],"pair":[57],"optimize":[59],"model":[61],"with":[62,171],"reconstruction":[63,160],"loss.":[64],"If":[65],"only":[67],"have":[68],"without":[70],"transcription,":[71],"generates":[73],"most":[75],"likely":[76],"transcription":[77,88],"from":[78,173],"data,":[81],"then":[83],"uses":[85],"generated":[87],"reconstruct":[90],"original":[92],"features.":[94],"However,":[95],"in":[96],"papers,":[98],"just":[100],"limited":[101],"back-propagation":[103],"closest":[106],"module,":[107],"which":[108],"is":[109,115,123],"part.":[112],"One":[113],"reason":[114],"that":[116],"back-propagating":[117],"error":[119],"through":[120],"challenging":[124],"due":[125],"output":[128],"of":[129],"being":[132],"discrete":[133],"tokens,":[134],"creating":[135],"non-differentiability":[136],"between":[137],"ASR.":[141],"this":[143,147],"paper,":[144],"address":[146],"problem":[148],"describe":[150],"how":[151],"thoroughly":[153],"train":[154],"end-to-end":[158],"loss":[161],"using":[162],"straight-through":[164],"estimator":[165],"(ST).":[166],"Experimental":[167],"results":[168],"revealed":[169],"that,":[170],"sampling":[172],"ST-Gumbel-Softmax,":[174],"were":[176],"able":[177],"update":[179],"parameters":[181],"improve":[183],"performances":[186],"by":[187],"11%":[188],"relative":[189],"CER":[190],"reduction":[191],"compared":[192],"baseline.":[195]},"counts_by_year":[{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":4},{"year":2021,"cited_by_count":5},{"year":2020,"cited_by_count":11},{"year":2019,"cited_by_count":8}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}