{"id":"https://openalex.org/W2936012302","doi":"https://doi.org/10.1109/icassp.2019.8682888","title":"Distributed Deep Learning Strategies for Automatic Speech Recognition","display_name":"Distributed Deep Learning Strategies for Automatic Speech Recognition","publication_year":2019,"publication_date":"2019-04-17","ids":{"openalex":"https://openalex.org/W2936012302","doi":"https://doi.org/10.1109/icassp.2019.8682888","mag":"2936012302"},"language":"en","primary_location":{"id":"doi:10.1109/icassp.2019.8682888","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2019.8682888","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":"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/A5014921180","display_name":"Wei Zhang","orcid":"https://orcid.org/0000-0002-6921-4926"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei Zhang","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102014291","display_name":"Xiaodong Cui","orcid":"https://orcid.org/0000-0003-4865-1307"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaodong Cui","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000286196","display_name":"Ulrich Finkler","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ulrich Finkler","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003725957","display_name":"Brian Kingsbury","orcid":"https://orcid.org/0000-0002-1343-6837"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Brian Kingsbury","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079994647","display_name":"George Saon","orcid":"https://orcid.org/0009-0004-6837-5009"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"George Saon","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063522852","display_name":"David S. Kung","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"David Kung","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5034529775","display_name":"Michael Picheny","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Michael Picheny","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":33,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"5706","last_page":"5710"},"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":0.9997000098228455,"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":0.8711588382720947},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.7353420257568359},{"id":"https://openalex.org/keywords/asynchronous-communication","display_name":"Asynchronous communication","score":0.6961126923561096},{"id":"https://openalex.org/keywords/test-set","display_name":"Test set","score":0.6134961843490601},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5331795811653137},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4874558448791504},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.47827231884002686},{"id":"https://openalex.org/keywords/sync","display_name":"sync","score":0.4633166790008545},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.44077977538108826},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.4368036389350891},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.4168388247489929},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.23974964022636414},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.12619972229003906},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.10714989900588989}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8711588382720947},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.7353420257568359},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.6961126923561096},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.6134961843490601},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5331795811653137},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4874558448791504},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.47827231884002686},{"id":"https://openalex.org/C3913047","wikidata":"https://www.wikidata.org/wiki/Q1956265","display_name":"sync","level":3,"score":0.4633166790008545},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44077977538108826},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.4368036389350891},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.4168388247489929},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.23974964022636414},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.12619972229003906},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.10714989900588989},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.0},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp.2019.8682888","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2019.8682888","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":46,"referenced_works":["https://openalex.org/W1825216778","https://openalex.org/W2022185456","https://openalex.org/W2057332538","https://openalex.org/W2160815625","https://openalex.org/W2168231600","https://openalex.org/W2173213060","https://openalex.org/W2184045248","https://openalex.org/W2193413348","https://openalex.org/W2336650964","https://openalex.org/W2405883473","https://openalex.org/W2407022425","https://openalex.org/W2412976325","https://openalex.org/W2557113115","https://openalex.org/W2568564101","https://openalex.org/W2617766261","https://openalex.org/W2622263826","https://openalex.org/W2623666582","https://openalex.org/W2749988060","https://openalex.org/W2753217274","https://openalex.org/W2759071281","https://openalex.org/W2775776326","https://openalex.org/W2884711234","https://openalex.org/W2885142640","https://openalex.org/W2889797931","https://openalex.org/W2952754453","https://openalex.org/W2953235111","https://openalex.org/W2962786385","https://openalex.org/W2962863496","https://openalex.org/W2962950660","https://openalex.org/W2963228337","https://openalex.org/W2963266252","https://openalex.org/W2963807318","https://openalex.org/W2963964896","https://openalex.org/W2964004663","https://openalex.org/W6684859321","https://openalex.org/W6687566353","https://openalex.org/W6692488971","https://openalex.org/W6703420464","https://openalex.org/W6738460352","https://openalex.org/W6739622702","https://openalex.org/W6743289643","https://openalex.org/W6745458143","https://openalex.org/W6746719585","https://openalex.org/W6753209298","https://openalex.org/W6753770798","https://openalex.org/W6780043775"],"related_works":["https://openalex.org/W4392306433","https://openalex.org/W2524993630","https://openalex.org/W2063547430","https://openalex.org/W2607244263","https://openalex.org/W2234430299","https://openalex.org/W1831690014","https://openalex.org/W1561022130","https://openalex.org/W1878822803","https://openalex.org/W1995263003","https://openalex.org/W3015584167"],"abstract_inverted_index":{"In":[0],"this":[1],"paper,":[2],"we":[3,104,143],"propose":[4],"and":[5,19,89,132],"investigate":[6,51],"a":[7,23,122,133],"variety":[8],"of":[9,39,92],"distributed":[10,78],"deep":[11],"learning":[12,58],"strategies":[13],"for":[14,45],"automatic":[15],"speech":[16],"recognition":[17],"(ASR)":[18],"evaluate":[20],"them":[21],"with":[22,64,115,153],"state-of-the-art":[24],"Long":[25],"short-term":[26],"memory":[27],"(LSTM)":[28],"acoustic":[29],"model":[30,72,109,147],"on":[31,125,136],"the":[32,40,54,62,71,90,93,107,126,137,146],"2000-hour":[33],"Switchboard":[34,128],"(SWB2000),":[35],"which":[36],"is":[37],"one":[38],"most":[41],"widely":[42],"used":[43],"datasets":[44],"ASR":[46],"performance":[47],"benchmark.":[48],"We":[49,74,101],"first":[50],"what":[52],"are":[53],"proper":[55],"hyper-parameters":[56],"(e.g.,":[57],"rate)":[59],"to":[60,96,120],"enable":[61],"training":[63],"sufficiently":[65],"large":[66],"batch":[67],"size":[68],"without":[69,158],"impairing":[70],"accuracy.":[73,161],"then":[75],"implement":[76],"various":[77],"strategies,":[79],"including":[80],"Synchronous":[81],"(SYNC)":[82],",":[83],"Asynchronous":[84],"Decentralized":[85],"Parallel":[86],"SGD":[87],"(ADPSGD)":[88],"hybrid":[91],"two":[94],"HYBRID,":[95],"study":[97],"their":[98],"runtime/accuracy":[99],"trade-off.":[100],"show":[102],"that":[103],"can":[105,144],"train":[106,145],"LSTM":[108],"using":[110,148],"ADPSGD":[111],"in":[112,150,160],"14":[113],"hours":[114,152],"16":[116],"NVIDIA":[117,155],"P100":[118],"GPUs":[119,157],"reach":[121],"7.6%":[123],"WER":[124,135],"Hub5-2000":[127],"(SWB)":[129],"test":[130,140],"set":[131],"13.1%":[134],"Call-Home":[138],"(CH)":[139],"set.":[141],"Furthermore,":[142],"HYBRID":[149],"11.5":[151],"32":[154],"V100":[156],"loss":[159]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":11},{"year":2020,"cited_by_count":12},{"year":2019,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}