{"id":"https://openalex.org/W2944311603","doi":"https://doi.org/10.1109/iscslp.2018.8706623","title":"Subspace Based Sequence Discriminative Training of LSTM Acoustic Models with Feed-Forward Layers","display_name":"Subspace Based Sequence Discriminative Training of LSTM Acoustic Models with Feed-Forward Layers","publication_year":2018,"publication_date":"2018-11-01","ids":{"openalex":"https://openalex.org/W2944311603","doi":"https://doi.org/10.1109/iscslp.2018.8706623","mag":"2944311603"},"language":"en","primary_location":{"id":"doi:10.1109/iscslp.2018.8706623","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iscslp.2018.8706623","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 11th International Symposium on Chinese Spoken Language Processing (ISCSLP)","raw_type":"proceedings-article"},"type":"article","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/A5004424826","display_name":"Lahiru Samarakoon","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090344","display_name":"Fano Labs (China)","ror":"https://ror.org/008jsgk36","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210090344"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Lahiru Samarakoon","raw_affiliation_strings":["Fano Labs, Hong Kong"],"affiliations":[{"raw_affiliation_string":"Fano Labs, Hong Kong","institution_ids":["https://openalex.org/I4210090344"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059141717","display_name":"Brian Mak","orcid":"https://orcid.org/0000-0001-6787-5555"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Brian Mak","raw_affiliation_strings":["Hong Kong University of Science and Technology"],"affiliations":[{"raw_affiliation_string":"Hong Kong University of Science and Technology","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5005693569","display_name":"Yuet-Ming Lam","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090344","display_name":"Fano Labs (China)","ror":"https://ror.org/008jsgk36","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210090344"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Y.S. Albert Lam","raw_affiliation_strings":["Fano Labs, Hong Kong"],"affiliations":[{"raw_affiliation_string":"Fano Labs, Hong Kong","institution_ids":["https://openalex.org/I4210090344"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5004424826"],"corresponding_institution_ids":["https://openalex.org/I4210090344"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.21439574,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"88","issue":null,"first_page":"136","last_page":"140"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":1.0,"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":1.0,"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/T10860","display_name":"Speech and Audio Processing","score":0.9998000264167786,"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/T11309","display_name":"Music and Audio Processing","score":0.9998000264167786,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.8510854244232178},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7307503819465637},{"id":"https://openalex.org/keywords/overfitting","display_name":"Overfitting","score":0.7107618451118469},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.56404709815979},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5552777051925659},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.48570212721824646},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.4828813374042511},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4769669771194458},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.4551250636577606},{"id":"https://openalex.org/keywords/subspace-topology","display_name":"Subspace topology","score":0.43145740032196045},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3220704197883606}],"concepts":[{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.8510854244232178},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7307503819465637},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.7107618451118469},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.56404709815979},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5552777051925659},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.48570212721824646},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.4828813374042511},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4769669771194458},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.4551250636577606},{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.43145740032196045},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3220704197883606},{"id":"https://openalex.org/C54355233","wikidata":"https://www.wikidata.org/wiki/Q7162","display_name":"Genetics","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}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/iscslp.2018.8706623","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iscslp.2018.8706623","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 11th International Symposium on Chinese Spoken Language Processing (ISCSLP)","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-95384","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-95384","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"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":"Conference paper"},{"id":"pmh:oai:repository.ust.hk:1783.1-95384","is_oa":false,"landing_page_url":"http://repository.ust.hk/ir/Record/1783.1-95384","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Conference paper"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","score":0.7400000095367432,"display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":38,"referenced_works":["https://openalex.org/W330298975","https://openalex.org/W1524333225","https://openalex.org/W1569447338","https://openalex.org/W1574530145","https://openalex.org/W1600744878","https://openalex.org/W1987238397","https://openalex.org/W1993882792","https://openalex.org/W2064675550","https://openalex.org/W2083751884","https://openalex.org/W2106554350","https://openalex.org/W2114016253","https://openalex.org/W2125234026","https://openalex.org/W2131342762","https://openalex.org/W2136848157","https://openalex.org/W2147768505","https://openalex.org/W2160306971","https://openalex.org/W2160815625","https://openalex.org/W2181607856","https://openalex.org/W2293009711","https://openalex.org/W2293634267","https://openalex.org/W2296748324","https://openalex.org/W2403195671","https://openalex.org/W2508162385","https://openalex.org/W2514741789","https://openalex.org/W2616139854","https://openalex.org/W2746465152","https://openalex.org/W2766238266","https://openalex.org/W2964084166","https://openalex.org/W4256161595","https://openalex.org/W4285719527","https://openalex.org/W6631362777","https://openalex.org/W6633847657","https://openalex.org/W6679429981","https://openalex.org/W6686045668","https://openalex.org/W6696934422","https://openalex.org/W6696982659","https://openalex.org/W6713280142","https://openalex.org/W6738021889"],"related_works":["https://openalex.org/W4362597605","https://openalex.org/W1574414179","https://openalex.org/W4297676672","https://openalex.org/W3009056573","https://openalex.org/W2922073769","https://openalex.org/W4281702477","https://openalex.org/W2490526372","https://openalex.org/W4376166922","https://openalex.org/W4378510483","https://openalex.org/W3040691452"],"abstract_inverted_index":{"State-of-the-art":[0],"automatic":[1],"speech":[2],"recognition":[3],"(ASR)":[4],"systems":[5,41],"use":[6],"sequence":[7,20,90],"discriminative":[8,21,91],"training":[9,22,60,92],"for":[10,137],"improved":[11],"performance":[12,45,88],"over":[13,166],"frame-level":[14],"cross-entropy":[15],"(CE)":[16],"criterion.":[17],"Even":[18],"though":[19],"improves":[23],"long":[24],"short-term":[25],"memory":[26],"(LSTM)":[27],"recurrent":[28],"neural":[29],"network":[30],"(RNN)":[31],"acoustic":[32],"models":[33],"(AMs),":[34],"it":[35],"is":[36,78,126],"not":[37,79],"clear":[38],"whether":[39],"these":[40],"achieve":[42],"the":[43,52,67,87,104,109,138,142,159],"optimal":[44],"due":[46],"to":[47,85],"overfitting.":[48],"This":[49],"paper":[50],"investigates":[51],"effect":[53],"of":[54,70,89,93,131,141],"state-level":[55],"minimum":[56],"Bayes":[57],"risk":[58],"(sMBR)":[59],"on":[61],"LSTM":[62,76,94,106,139,171],"AMs":[63],"and":[64,108],"shows":[65],"that":[66,158],"conventional":[68],"way":[69],"performing":[71,135],"sMBR":[72,121,136,169],"by":[73],"updating":[74],"all":[75],"parameters":[77],"optimal.":[80],"We":[81,156],"investigate":[82],"two":[83],"methods":[84,145],"improve":[86],"AMs.":[95],"First":[96],"more":[97,119],"feed-forward":[98],"(FF)":[99],"layers":[100,116,140],"are":[101,146],"included":[102],"between":[103],"last":[105],"layer":[107,111],"output":[110],"so":[112],"those":[113],"additional":[114],"FF":[115],"may":[117],"benefit":[118],"from":[120],"training.":[122],"Second,":[123],"a":[124,167],"subspace":[125],"estimated":[127],"as":[128],"an":[129],"interpolation":[130],"rank-1":[132],"matrices":[133],"when":[134],"AM.":[143],"Our":[144],"evaluated":[147],"in":[148],"benchmark":[149],"AMI":[150],"single":[151],"distance":[152],"microphone":[153],"(SDM)":[154],"task.":[155],"find":[157],"proposed":[160],"approaches":[161],"provide":[162],"1.6%":[163],"absolute":[164],"improvement":[165],"strong":[168],"trained":[170],"baseline.":[172]},"counts_by_year":[],"updated_date":"2026-04-15T08:11:43.952461","created_date":"2025-10-10T00:00:00"}