{"id":"https://openalex.org/W2034261344","doi":"https://doi.org/10.1109/slt.2014.7078571","title":"A distributed architecture for fast SGD sequence discriminative training of DNN acoustic models","display_name":"A distributed architecture for fast SGD sequence discriminative training of DNN acoustic models","publication_year":2014,"publication_date":"2014-12-01","ids":{"openalex":"https://openalex.org/W2034261344","doi":"https://doi.org/10.1109/slt.2014.7078571","mag":"2034261344"},"language":"en","primary_location":{"id":"doi:10.1109/slt.2014.7078571","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt.2014.7078571","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2014 IEEE Spoken Language Technology Workshop (SLT)","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/A5079994647","display_name":"George Saon","orcid":"https://orcid.org/0009-0004-6837-5009"},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"George Saon","raw_affiliation_strings":["IBM T. J. Watson Research Center, NY, USA","[IBM T.J. Watson Research Center, Yorktown Heights, NY 10598, USA]"],"affiliations":[{"raw_affiliation_string":"IBM T. J. Watson Research Center, NY, USA","institution_ids":[]},{"raw_affiliation_string":"[IBM T.J. Watson Research Center, Yorktown Heights, NY 10598, USA]","institution_ids":["https://openalex.org/I1341412227"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5079994647"],"corresponding_institution_ids":["https://openalex.org/I1341412227"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.08049439,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"183","last_page":"188"},"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/T11309","display_name":"Music and Audio Processing","score":0.9998999834060669,"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/T10860","display_name":"Speech and Audio Processing","score":0.9995999932289124,"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/computer-science","display_name":"Computer science","score":0.8393494486808777},{"id":"https://openalex.org/keywords/queue","display_name":"Queue","score":0.5078089833259583},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.42920345067977905},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.4136284291744232},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.3879636228084564},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2902774214744568},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.1754302978515625}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8393494486808777},{"id":"https://openalex.org/C160403385","wikidata":"https://www.wikidata.org/wiki/Q220543","display_name":"Queue","level":2,"score":0.5078089833259583},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.42920345067977905},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4136284291744232},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3879636228084564},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2902774214744568},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.1754302978515625},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/slt.2014.7078571","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt.2014.7078571","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2014 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Reduced inequalities","score":0.7300000190734863,"id":"https://metadata.un.org/sdg/10"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320306078","display_name":"U.S. Department of Defense","ror":"https://ror.org/0447fe631"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W1985371235","https://openalex.org/W1987238397","https://openalex.org/W2028706510","https://openalex.org/W2035846950","https://openalex.org/W2058641082","https://openalex.org/W2079623482","https://openalex.org/W2087402357","https://openalex.org/W2094233035","https://openalex.org/W2114016253","https://openalex.org/W2131342762","https://openalex.org/W2160306971","https://openalex.org/W2160815625","https://openalex.org/W2168231600","https://openalex.org/W2294723541","https://openalex.org/W2403195671","https://openalex.org/W2405787239","https://openalex.org/W6679429981","https://openalex.org/W6684859321","https://openalex.org/W6697185088","https://openalex.org/W6713280142","https://openalex.org/W6713989789"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2965546495","https://openalex.org/W4389116644","https://openalex.org/W2153315159","https://openalex.org/W3103844505","https://openalex.org/W259157601","https://openalex.org/W4205463238","https://openalex.org/W2761785940","https://openalex.org/W2110523656"],"abstract_inverted_index":{"We":[0],"describe":[1],"a":[2,17,34,47,63,88],"hybrid":[3],"GPU/CPU":[4],"architecture":[5],"for":[6],"stochastic":[7],"gradient":[8],"descent":[9],"training":[10,108],"of":[11,26,49,91,103,111,116,125],"neural":[12],"network":[13],"acoustic":[14],"models":[15,77],"under":[16],"lattice-based":[18],"minimum":[19],"Bayes":[20],"risk":[21],"(MBR)":[22],"criterion.":[23],"The":[24],"crux":[25],"the":[27,68,79,84,101],"method":[28],"is":[29,97],"to":[30,78,99],"run":[31],"SGD":[32,85,105],"on":[33,46,67,118,133],"GPU":[35,69],"card":[36],"which":[37,53],"consumes":[38],"frame-randomized":[39],"mini-batches":[40],"produced":[41],"by":[42],"multiple":[43],"workers":[44],"running":[45,66],"cluster":[48],"multi-core":[50],"CPU":[51],"nodes":[52],"compute":[54],"HMM":[55],"state":[56],"MBR":[57],"occupancies.":[58],"To":[59],"minimize":[60],"communication":[61],"cost,":[62],"separate":[64],"thread":[65,86],"host":[70],"receives":[71],"minibatches":[72],"from":[73],"and":[74,81,129,145],"sends":[75],"updated":[76],"workers,":[80],"communicates":[82],"with":[83],"via":[87],"producer-consumer":[89],"queue":[90],"minibatches.":[92],"Using":[93],"this":[94],"architecture,":[95],"it":[96],"possible":[98],"match":[100],"speed":[102],"GPU-based":[104],"cross-entropy":[106],"(CE)":[107],"(1":[109],"hour":[110],"processing":[112],"per":[113],"100":[114],"hours":[115],"audio":[117],"Switchboard).":[119],"Additionally,":[120],"we":[121],"compare":[122],"different":[123],"ways":[124],"doing":[126],"frame":[127],"randomization":[128],"discuss":[130],"experimental":[131],"results":[132],"three":[134],"LVCSR":[135],"tasks":[136],"(Switchboard":[137],"300":[138,150],"hours,":[139,144],"English":[140],"broadcast":[141],"news":[142],"50":[143],"noisy":[146],"Levantine":[147],"telephone":[148],"conversations":[149],"hours).":[151]},"counts_by_year":[],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}