{"id":"https://openalex.org/W2892124901","doi":"https://doi.org/10.1109/icassp.2018.8462245","title":"SEQUENCE TRAINING OF ENCODER-DECODER MODEL USING POLICY GRADIENT FOR END- TO-END SPEECH RECOGNITION","display_name":"SEQUENCE TRAINING OF ENCODER-DECODER MODEL USING POLICY GRADIENT FOR END- TO-END SPEECH RECOGNITION","publication_year":2018,"publication_date":"2018-04-01","ids":{"openalex":"https://openalex.org/W2892124901","doi":"https://doi.org/10.1109/icassp.2018.8462245","mag":"2892124901"},"language":"en","primary_location":{"id":"doi:10.1109/icassp.2018.8462245","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2018.8462245","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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/A5036532214","display_name":"Shigeki Karita","orcid":null},"institutions":[{"id":"https://openalex.org/I2251713219","display_name":"NTT (Japan)","ror":"https://ror.org/00berct97","country_code":"JP","type":"company","lineage":["https://openalex.org/I2251713219"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Shigeki Karita","raw_affiliation_strings":["NTT Communication Science Laboratories, NTT Corporation, Kyoto, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NTT Communication Science Laboratories, NTT Corporation, Kyoto, Japan","institution_ids":["https://openalex.org/I2251713219"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101949937","display_name":"Atsunori Ogawa","orcid":"https://orcid.org/0000-0002-2888-101X"},"institutions":[{"id":"https://openalex.org/I2251713219","display_name":"NTT (Japan)","ror":"https://ror.org/00berct97","country_code":"JP","type":"company","lineage":["https://openalex.org/I2251713219"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Atsunori Ogawa","raw_affiliation_strings":["NTT Communication Science Laboratories, NTT Corporation, Kyoto, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NTT Communication Science Laboratories, NTT Corporation, Kyoto, Japan","institution_ids":["https://openalex.org/I2251713219"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023868166","display_name":"Marc Delcroix","orcid":"https://orcid.org/0000-0002-5175-7834"},"institutions":[{"id":"https://openalex.org/I2251713219","display_name":"NTT (Japan)","ror":"https://ror.org/00berct97","country_code":"JP","type":"company","lineage":["https://openalex.org/I2251713219"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Marc Delcroix","raw_affiliation_strings":["NTT Communication Science Laboratories, NTT Corporation, Kyoto, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NTT Communication Science Laboratories, NTT Corporation, Kyoto, Japan","institution_ids":["https://openalex.org/I2251713219"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5021240106","display_name":"Tomohiro Nakatani","orcid":"https://orcid.org/0000-0002-7487-7150"},"institutions":[{"id":"https://openalex.org/I2251713219","display_name":"NTT (Japan)","ror":"https://ror.org/00berct97","country_code":"JP","type":"company","lineage":["https://openalex.org/I2251713219"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Tomohiro Nakatani","raw_affiliation_strings":["NTT Communication Science Laboratories, NTT Corporation, Kyoto, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NTT Communication Science Laboratories, NTT Corporation, Kyoto, Japan","institution_ids":["https://openalex.org/I2251713219"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I2251713219"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":19,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9995999932289124,"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":0.9995999932289124,"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":0.9994999766349792,"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":0.9939000010490417,"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":0.8054382801055908},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.7086743116378784},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.7075673341751099},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.6766728758811951},{"id":"https://openalex.org/keywords/soft-decision-decoder","display_name":"Soft-decision decoder","score":0.6549268364906311},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.645578920841217},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5617987513542175},{"id":"https://openalex.org/keywords/ground-truth","display_name":"Ground truth","score":0.5522373914718628},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.5258356332778931},{"id":"https://openalex.org/keywords/cross-entropy","display_name":"Cross entropy","score":0.4881230294704437},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.47303932905197144},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.4354918897151947},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.42298033833503723},{"id":"https://openalex.org/keywords/beam-search","display_name":"Beam search","score":0.4145796298980713},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.412355899810791},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3920442461967468},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3892669379711151},{"id":"https://openalex.org/keywords/search-algorithm","display_name":"Search algorithm","score":0.1019631028175354},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.06689050793647766}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8054382801055908},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.7086743116378784},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.7075673341751099},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.6766728758811951},{"id":"https://openalex.org/C185588885","wikidata":"https://www.wikidata.org/wiki/Q7553811","display_name":"Soft-decision decoder","level":3,"score":0.6549268364906311},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.645578920841217},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5617987513542175},{"id":"https://openalex.org/C146849305","wikidata":"https://www.wikidata.org/wiki/Q370766","display_name":"Ground truth","level":2,"score":0.5522373914718628},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.5258356332778931},{"id":"https://openalex.org/C167981619","wikidata":"https://www.wikidata.org/wiki/Q1685498","display_name":"Cross entropy","level":3,"score":0.4881230294704437},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.47303932905197144},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.4354918897151947},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42298033833503723},{"id":"https://openalex.org/C19889080","wikidata":"https://www.wikidata.org/wiki/Q2835852","display_name":"Beam search","level":3,"score":0.4145796298980713},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.412355899810791},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3920442461967468},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3892669379711151},{"id":"https://openalex.org/C125583679","wikidata":"https://www.wikidata.org/wiki/Q755673","display_name":"Search algorithm","level":2,"score":0.1019631028175354},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.06689050793647766},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C54355233","wikidata":"https://www.wikidata.org/wiki/Q7162","display_name":"Genetics","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},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp.2018.8462245","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2018.8462245","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 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":0.7300000190734863}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W38527073","https://openalex.org/W811578723","https://openalex.org/W1514535095","https://openalex.org/W1522301498","https://openalex.org/W1524333225","https://openalex.org/W1533416326","https://openalex.org/W1646152356","https://openalex.org/W2066378046","https://openalex.org/W2109886035","https://openalex.org/W2130942839","https://openalex.org/W2143612262","https://openalex.org/W2155027007","https://openalex.org/W2176263492","https://openalex.org/W2180816288","https://openalex.org/W2250357346","https://openalex.org/W2257979135","https://openalex.org/W2526425061","https://openalex.org/W2530876040","https://openalex.org/W2749475572","https://openalex.org/W2962765220","https://openalex.org/W2964352247","https://openalex.org/W6601546718","https://openalex.org/W6622796164","https://openalex.org/W6630875275","https://openalex.org/W6631190155","https://openalex.org/W6631362777","https://openalex.org/W6631839158","https://openalex.org/W6636774829","https://openalex.org/W6667030123","https://openalex.org/W6676562027","https://openalex.org/W6679436768","https://openalex.org/W6683204974","https://openalex.org/W6685322675","https://openalex.org/W6686002079","https://openalex.org/W6691509046","https://openalex.org/W6722524744","https://openalex.org/W6728030952","https://openalex.org/W6728892627","https://openalex.org/W6738974909","https://openalex.org/W6743283956"],"related_works":["https://openalex.org/W1566315437","https://openalex.org/W2114916462","https://openalex.org/W2594897229","https://openalex.org/W4221142855","https://openalex.org/W2151348424","https://openalex.org/W2050138804","https://openalex.org/W71447404","https://openalex.org/W4290708361","https://openalex.org/W2129812225","https://openalex.org/W2523799048"],"abstract_inverted_index":{"The":[0],"standard":[1],"evaluation":[2,94],"metric":[3],"of":[4,123,133,161,176,185],"automatic":[5],"speech":[6],"recognition":[7,183],"(ASR)":[8],"is":[9,99],"the":[10,17,67,87,97,112,120,124,131,138,148,158,162,169,174,182,186],"word":[11,21],"error":[12],"rate":[13],"(WER),":[14],"which":[15,115],"measures":[16],"dissimilarity":[18],"between":[19],"recognized":[20],"sequences":[22],"and":[23],"their":[24],"ground":[25,149],"truth.":[26,150],"Many":[27],"training":[28,128,145],"algorithms":[29],"designed":[30],"to":[31,110,118,156,172],"reduce":[32],"sequence-level":[33,93],"errors":[34],"such":[35],"as":[36,136],"WER":[37,98,122],"have":[38],"been":[39],"proposed":[40,178],"for":[41,60,85,180],"hidden":[42],"Markov":[43],"model":[44,62,69,89,125],"(HMM)-based":[45],"ASR,":[46,65],"e.g.,":[47],"state-level":[48],"minimum":[49],"Bayes":[50],"risk":[51],"(sMBR).":[52],"However,":[53],"these":[54],"approaches":[55],"cannot":[56],"be":[57],"used":[58],"directly":[59,101],"encoder-decoder":[61,68,88,113,163,187],"based":[63,90],"end-to-end":[64],"because":[66],"employs":[70,130],"very":[71],"different":[72],"mechanisms":[73],"from":[74],"HMM-based":[75],"approaches.":[76],"In":[77],"this":[78],"paper,":[79],"we":[80,103,152],"propose":[81],"a":[82,92,105],"new":[83],"method":[84,129,179],"optimizing":[86],"on":[91],"metric.":[95],"Since":[96],"not":[100],"differentiable,":[102],"adopt":[104],"policy":[106],"gradient":[107],"objective":[108],"function":[109],"train":[111],"model,":[114],"enables":[116],"us":[117],"minimize":[119],"expected":[121],"predictions.":[126],"This":[127],"scoring":[132],"multiple":[134],"hypotheses":[135],"in":[137],"decoding":[139,159],"stage":[140],"while":[141],"usual":[142],"cross":[143],"entropy":[144],"uses":[146],"only":[147],"Therefore,":[151],"can":[153],"expect":[154],"it":[155],"improve":[157],"results":[160],"model.":[164,188],"We":[165],"perform":[166],"experiments":[167],"using":[168],"Tedlium":[170],"corpus":[171],"demonstrate":[173],"potential":[175],"our":[177],"improving":[181],"performance":[184]},"counts_by_year":[{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":1},{"year":2020,"cited_by_count":2},{"year":2019,"cited_by_count":8},{"year":2018,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}