{"id":"https://openalex.org/W3161768215","doi":"https://doi.org/10.1109/icassp39728.2021.9414087","title":"Transformer in Action: A Comparative Study of Transformer-Based Acoustic Models for Large Scale Speech Recognition Applications","display_name":"Transformer in Action: A Comparative Study of Transformer-Based Acoustic Models for Large Scale Speech Recognition Applications","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3161768215","doi":"https://doi.org/10.1109/icassp39728.2021.9414087","mag":"3161768215"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9414087","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414087","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 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/A5100339989","display_name":"Yongqiang Wang","orcid":"https://orcid.org/0000-0002-8600-0276"},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yongqiang Wang","raw_affiliation_strings":["Facebook AI, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103247973","display_name":"Yangyang Shi","orcid":"https://orcid.org/0000-0001-5297-4155"},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yangyang Shi","raw_affiliation_strings":["Facebook AI, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101984914","display_name":"Frank Zhang","orcid":"https://orcid.org/0000-0001-9946-6502"},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Frank Zhang","raw_affiliation_strings":["Facebook AI, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101579713","display_name":"Chunyang Wu","orcid":"https://orcid.org/0000-0002-0269-3555"},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chunyang Wu","raw_affiliation_strings":["Facebook AI, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051075420","display_name":"Julian Chan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Julian Chan","raw_affiliation_strings":["Facebook AI, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087421101","display_name":"Ching-Feng Yeh","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ching-Feng Yeh","raw_affiliation_strings":["Facebook AI, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5056584611","display_name":"Alex Xiao","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alex Xiao","raw_affiliation_strings":["Facebook AI, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210114444"],"apc_list":null,"apc_paid":null,"fwci":2.0514,"has_fulltext":false,"cited_by_count":22,"citation_normalized_percentile":{"value":0.89249825,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"6778","last_page":"6782"},"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/T11309","display_name":"Music and Audio Processing","score":0.9994999766349792,"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":0.9980999827384949,"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":0.7620837688446045},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.7424257397651672},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6587003469467163},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6458200216293335},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5948213338851929},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.5378962755203247},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.4371682405471802},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3173413574695587},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.2865116596221924},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.12233293056488037},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.10321652889251709},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.09585845470428467},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.0705992579460144}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7620837688446045},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7424257397651672},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6587003469467163},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6458200216293335},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5948213338851929},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.5378962755203247},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.4371682405471802},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3173413574695587},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.2865116596221924},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.12233293056488037},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.10321652889251709},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.09585845470428467},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0705992579460144}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9414087","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414087","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":55,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W1524333225","https://openalex.org/W1553004968","https://openalex.org/W1828163288","https://openalex.org/W1920942766","https://openalex.org/W1995562189","https://openalex.org/W2097268427","https://openalex.org/W2114016253","https://openalex.org/W2121879602","https://openalex.org/W2127141656","https://openalex.org/W2133564696","https://openalex.org/W2160815625","https://openalex.org/W2184045248","https://openalex.org/W2293634267","https://openalex.org/W2394932179","https://openalex.org/W2402146185","https://openalex.org/W2407080277","https://openalex.org/W2896457183","https://openalex.org/W2936774411","https://openalex.org/W2963341956","https://openalex.org/W2963403868","https://openalex.org/W2964084166","https://openalex.org/W2964121744","https://openalex.org/W2964308564","https://openalex.org/W2972818416","https://openalex.org/W2981857663","https://openalex.org/W3007328579","https://openalex.org/W3008525923","https://openalex.org/W3015960524","https://openalex.org/W3015974384","https://openalex.org/W3016010032","https://openalex.org/W3025165719","https://openalex.org/W3028274012","https://openalex.org/W3096888553","https://openalex.org/W3097777922","https://openalex.org/W3144557079","https://openalex.org/W3146288864","https://openalex.org/W3146505093","https://openalex.org/W4385245566","https://openalex.org/W6631190155","https://openalex.org/W6631362777","https://openalex.org/W6638749077","https://openalex.org/W6640172592","https://openalex.org/W6679434410","https://openalex.org/W6696934422","https://openalex.org/W6711962127","https://openalex.org/W6712930963","https://openalex.org/W6713762819","https://openalex.org/W6739901393","https://openalex.org/W6755207826","https://openalex.org/W6777615951","https://openalex.org/W6780226713","https://openalex.org/W6785222072","https://openalex.org/W6785229415","https://openalex.org/W6786215725"],"related_works":["https://openalex.org/W3008625068","https://openalex.org/W3128807919","https://openalex.org/W3176411177","https://openalex.org/W4206178588","https://openalex.org/W4287635093","https://openalex.org/W3094491777","https://openalex.org/W3214715529","https://openalex.org/W3035501883","https://openalex.org/W3037446006","https://openalex.org/W4287752972"],"abstract_inverted_index":{"Transformer-based":[0],"acoustic":[1,24,38],"models":[2,39],"have":[3],"shown":[4],"promising":[5],"results":[6],"very":[7],"recently.":[8],"In":[9],"this":[10],"paper,":[11],"we":[12,49],"summarize":[13],"the":[14,35],"application":[15],"of":[16],"transformer":[17,36],"and":[18,60,95],"its":[19],"streamable":[20],"variant,":[21],"Emformer":[22,51,73,97],"based":[23,37],"model":[25,93],"[1]":[26],"for":[27],"large":[28],"scale":[29,46],"speech":[30],"recognition":[31],"applications.":[32],"We":[33],"compare":[34,50],"with":[40,52,89,91,108],"their":[41],"LSTM":[42,61],"counterparts":[43],"on":[44,56,62],"industrial":[45],"tasks.":[47,65],"Specifically,":[48],"latency-controlled":[53],"BLSTM":[54],"(LCBLSTM)":[55],"medium":[57,85],"latency":[58,64,69,86],"tasks":[59],"low":[63,68],"On":[66],"a":[67],"voice":[70],"assistant":[71],"task,":[72],"gets":[74,98],"24%":[75],"to":[76],"26%":[77],"relative":[78],"word":[79],"error":[80],"rate":[81],"reductions":[82],"(WERRs).":[83],"For":[84],"scenarios,":[87],"comparing":[88],"LCBLSTM":[90],"similar":[92],"size":[94],"latency,":[96],"significant":[99],"WERR":[100],"across":[101],"four":[102],"languages":[103],"in":[104],"video":[105],"captioning":[106],"datasets":[107],"2-3":[109],"times":[110],"inference":[111],"real-time":[112],"factors":[113],"reduction.":[114]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":7},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":4}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}