{"id":"https://openalex.org/W4372267411","doi":"https://doi.org/10.1109/icassp49357.2023.10096679","title":"Powerful and Extensible WFST Framework for Rnn-Transducer Losses","display_name":"Powerful and Extensible WFST Framework for Rnn-Transducer Losses","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372267411","doi":"https://doi.org/10.1109/icassp49357.2023.10096679"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10096679","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096679","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"preprint","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/A5089101632","display_name":"Aleksandr Laptev","orcid":"https://orcid.org/0000-0002-4690-705X"},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]},{"id":"https://openalex.org/I173089394","display_name":"ITMO University","ror":"https://ror.org/04txgxn49","country_code":"RU","type":"education","lineage":["https://openalex.org/I173089394"]}],"countries":["RU","US"],"is_corresponding":true,"raw_author_name":"Aleksandr Laptev","raw_affiliation_strings":["Nvidia,USA","Itmo University, St. Petersburg, Russia","Nvidia, USA"],"affiliations":[{"raw_affiliation_string":"Nvidia,USA","institution_ids":["https://openalex.org/I4210127875"]},{"raw_affiliation_string":"Itmo University, St. Petersburg, Russia","institution_ids":["https://openalex.org/I173089394"]},{"raw_affiliation_string":"Nvidia, USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103270162","display_name":"Vladimir Bataev","orcid":"https://orcid.org/0009-0005-7845-5042"},"institutions":[{"id":"https://openalex.org/I124357947","display_name":"University of London","ror":"https://ror.org/04cw6st05","country_code":"GB","type":"education","lineage":["https://openalex.org/I124357947"]},{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Vladimir Bataev","raw_affiliation_strings":["Nvidia,USA","Nvidia, USA","University of London, London, UK"],"affiliations":[{"raw_affiliation_string":"Nvidia,USA","institution_ids":["https://openalex.org/I4210127875"]},{"raw_affiliation_string":"Nvidia, USA","institution_ids":["https://openalex.org/I4210127875"]},{"raw_affiliation_string":"University of London, London, UK","institution_ids":["https://openalex.org/I124357947"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038012674","display_name":"Igor Gitman","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Igor Gitman","raw_affiliation_strings":["Nvidia,USA","Nvidia, USA"],"affiliations":[{"raw_affiliation_string":"Nvidia,USA","institution_ids":["https://openalex.org/I4210127875"]},{"raw_affiliation_string":"Nvidia, USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5032957280","display_name":"Boris Ginsburg","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Boris Ginsburg","raw_affiliation_strings":["Nvidia,USA","Nvidia, USA"],"affiliations":[{"raw_affiliation_string":"Nvidia,USA","institution_ids":["https://openalex.org/I4210127875"]},{"raw_affiliation_string":"Nvidia, USA","institution_ids":["https://openalex.org/I4210127875"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5089101632"],"corresponding_institution_ids":["https://openalex.org/I173089394","https://openalex.org/I4210127875"],"apc_list":null,"apc_paid":null,"fwci":0.5214,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.70246035,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9993000030517578,"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":0.9993000030517578,"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":0.9975000023841858,"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/T12031","display_name":"Speech and dialogue systems","score":0.9847999811172485,"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":0.8271050453186035},{"id":"https://openalex.org/keywords/transducer","display_name":"Transducer","score":0.6493295431137085},{"id":"https://openalex.org/keywords/debugging","display_name":"Debugging","score":0.5647947788238525},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.515414834022522},{"id":"https://openalex.org/keywords/python","display_name":"Python (programming language)","score":0.4911048412322998},{"id":"https://openalex.org/keywords/implementation","display_name":"Implementation","score":0.45603975653648376},{"id":"https://openalex.org/keywords/extensibility","display_name":"Extensibility","score":0.42566725611686707},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3506350517272949},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.309725284576416},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.13891175389289856}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8271050453186035},{"id":"https://openalex.org/C56318395","wikidata":"https://www.wikidata.org/wiki/Q215928","display_name":"Transducer","level":2,"score":0.6493295431137085},{"id":"https://openalex.org/C168065819","wikidata":"https://www.wikidata.org/wiki/Q845566","display_name":"Debugging","level":2,"score":0.5647947788238525},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.515414834022522},{"id":"https://openalex.org/C519991488","wikidata":"https://www.wikidata.org/wiki/Q28865","display_name":"Python (programming language)","level":2,"score":0.4911048412322998},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.45603975653648376},{"id":"https://openalex.org/C32833848","wikidata":"https://www.wikidata.org/wiki/Q4115054","display_name":"Extensibility","level":2,"score":0.42566725611686707},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3506350517272949},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.309725284576416},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.13891175389289856},{"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/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49357.2023.10096679","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096679","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 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":34,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1524333225","https://openalex.org/W1828163288","https://openalex.org/W2127141656","https://openalex.org/W2245493112","https://openalex.org/W2746192915","https://openalex.org/W2939297570","https://openalex.org/W2963211739","https://openalex.org/W2974231335","https://openalex.org/W3008174054","https://openalex.org/W3016234571","https://openalex.org/W3094979069","https://openalex.org/W3095697114","https://openalex.org/W3097777922","https://openalex.org/W3149509723","https://openalex.org/W3162833755","https://openalex.org/W3163907627","https://openalex.org/W3197956343","https://openalex.org/W3198643121","https://openalex.org/W3202184514","https://openalex.org/W3211040052","https://openalex.org/W4220743925","https://openalex.org/W4281952183","https://openalex.org/W4283700324","https://openalex.org/W4287647128","https://openalex.org/W4296069356","https://openalex.org/W4297841396","https://openalex.org/W6631362777","https://openalex.org/W6638749077","https://openalex.org/W6767671539","https://openalex.org/W6780226713","https://openalex.org/W6783314596","https://openalex.org/W6809993426","https://openalex.org/W6839217812"],"related_works":["https://openalex.org/W1948607442","https://openalex.org/W4321442002","https://openalex.org/W2015265939","https://openalex.org/W2284072287","https://openalex.org/W2611067230","https://openalex.org/W2480201319","https://openalex.org/W3004004161","https://openalex.org/W2387706296","https://openalex.org/W2155788121","https://openalex.org/W4214505573"],"abstract_inverted_index":{"This":[0],"paper":[1],"presents":[2],"a":[3,57,100,123],"framework":[4,147],"based":[5,55],"on":[6,56],"Weighted":[7],"Finite-State":[8],"Transducers":[9],"(WFST)":[10],"to":[11,31,38,73],"simplify":[12],"the":[13,60,79,93,105,108,119,133,145,152],"development":[14],"of":[15,23,59,95,99,107,130,137],"modifications":[16],"for":[17,82],"RNN-Transducer":[18],"(RNN-T)":[19],"loss.":[20],"Existing":[21],"implementations":[22],"RNN-T":[24,51,121,140],"use":[25],"CUDA-related":[26],"code,":[27],"which":[28,77],"is":[29],"hard":[30],"extend":[32],"and":[33,40,42,65,71,88,135,148],"debug.":[34],"WFSTs":[35],"are":[36,142,149],"easy":[37,72],"construct":[39],"extend,":[41],"allow":[43],"debugging":[44],"through":[45,97],"visualization.":[46],"We":[47,91],"introduce":[48],"two":[49],"WFST-powered":[50],"implementations:":[52],"(1)":[53],"\"Compose-Transducer\",":[54],"composition":[58],"WFST":[61],"graphs":[62],"from":[63],"acoustic":[64],"textual":[66],"schema":[67],"\u2013":[68,85,104],"computationally":[69,89],"competitive":[70],"modify;":[74],"(2)":[75],"\"Grid-Transducer\",":[76],"constructs":[78],"lattice":[80],"directly":[81],"further":[83],"computations":[84],"most":[86],"compact,":[87],"efficient.":[90],"illustrate":[92],"ease":[94],"extensibility":[96],"introduction":[98],"new":[101],"W-Transducer":[102,115],"loss":[103],"adaptation":[106],"Connectionist":[109],"Temporal":[110],"Classification":[111],"with":[112,127,144],"Wild":[113],"Cards.":[114],"(W-RNNT)":[116],"consistently":[117],"outperforms":[118],"standard":[120],"in":[122,151],"weakly-supervised":[124],"data":[125],"setup":[126],"missing":[128],"parts":[129],"transcriptions":[131],"at":[132],"beginning":[134],"end":[136],"utterances.":[138],"All":[139],"losses":[141],"implemented":[143],"k2":[146],"available":[150],"NeMo":[153],"toolkit.":[154]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2023,"cited_by_count":1}],"updated_date":"2026-04-03T22:45:19.894376","created_date":"2025-10-10T00:00:00"}