{"id":"https://openalex.org/W4200629210","doi":"https://doi.org/10.1109/icassp43922.2022.9746579","title":"Consistent Training and Decoding for End-to-End Speech Recognition Using Lattice-Free MMI","display_name":"Consistent Training and Decoding for End-to-End Speech Recognition Using Lattice-Free MMI","publication_year":2022,"publication_date":"2022-04-27","ids":{"openalex":"https://openalex.org/W4200629210","doi":"https://doi.org/10.1109/icassp43922.2022.9746579"},"language":"en","primary_location":{"id":"doi:10.1109/icassp43922.2022.9746579","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9746579","pdf_url":null,"source":{"id":"https://openalex.org/S4363607702","display_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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/A5068192693","display_name":"Jinchuan Tian","orcid":"https://orcid.org/0000-0002-2129-471X"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Jinchuan Tian","raw_affiliation_strings":["Peking University,ADSPLAB, School of ECE,Shenzhen,China","ADSPLAB, School of ECE, Peking University, Shenzhen, China"],"affiliations":[{"raw_affiliation_string":"Peking University,ADSPLAB, School of ECE,Shenzhen,China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"ADSPLAB, School of ECE, Peking University, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102286496","display_name":"Jianwei Yu","orcid":null},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jianwei Yu","raw_affiliation_strings":["Tencent AI Lab","Tencent ASR Oteam"],"affiliations":[{"raw_affiliation_string":"Tencent AI Lab","institution_ids":["https://openalex.org/I2250653659"]},{"raw_affiliation_string":"Tencent ASR Oteam","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106404246","display_name":"Chao Weng","orcid":null},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chao Weng","raw_affiliation_strings":["Tencent AI Lab","Tencent ASR Oteam"],"affiliations":[{"raw_affiliation_string":"Tencent AI Lab","institution_ids":["https://openalex.org/I2250653659"]},{"raw_affiliation_string":"Tencent ASR Oteam","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101785327","display_name":"Shi-Xiong Zhang","orcid":"https://orcid.org/0009-0007-7803-2407"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shi-Xiong Zhang","raw_affiliation_strings":["Tencent AI Lab"],"affiliations":[{"raw_affiliation_string":"Tencent AI Lab","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075183307","display_name":"Dan Su","orcid":"https://orcid.org/0000-0001-5746-9545"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dan Su","raw_affiliation_strings":["Tencent AI Lab"],"affiliations":[{"raw_affiliation_string":"Tencent AI Lab","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034476404","display_name":"Dong Yu","orcid":"https://orcid.org/0000-0003-0520-6844"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dong Yu","raw_affiliation_strings":["Tencent AI Lab"],"affiliations":[{"raw_affiliation_string":"Tencent AI Lab","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002795838","display_name":"Yuexian Zou","orcid":"https://orcid.org/0000-0001-9999-6140"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuexian Zou","raw_affiliation_strings":["Peking University,ADSPLAB, School of ECE,Shenzhen,China","ADSPLAB, School of ECE, Peking University, Shenzhen, China"],"affiliations":[{"raw_affiliation_string":"Peking University,ADSPLAB, School of ECE,Shenzhen,China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"ADSPLAB, School of ECE, Peking University, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5068192693"],"corresponding_institution_ids":["https://openalex.org/I20231570"],"apc_list":null,"apc_paid":null,"fwci":1.3527,"has_fulltext":false,"cited_by_count":13,"citation_normalized_percentile":{"value":0.8229812,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"7782","last_page":"7786"},"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":0.9990000128746033,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9988999962806702,"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/decoding-methods","display_name":"Decoding methods","score":0.7828823328018188},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.7299138307571411},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6908608078956604},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.6444820761680603},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.605404257774353},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5402249097824097},{"id":"https://openalex.org/keywords/mutual-information","display_name":"Mutual information","score":0.41767388582229614},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4122752547264099},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.4107138216495514},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.28042757511138916}],"concepts":[{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.7828823328018188},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.7299138307571411},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6908608078956604},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6444820761680603},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.605404257774353},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5402249097824097},{"id":"https://openalex.org/C152139883","wikidata":"https://www.wikidata.org/wiki/Q252973","display_name":"Mutual information","level":2,"score":0.41767388582229614},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4122752547264099},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.4107138216495514},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.28042757511138916},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp43922.2022.9746579","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9746579","pdf_url":null,"source":{"id":"https://openalex.org/S4363607702","display_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.6499999761581421}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W1828163288","https://openalex.org/W2114016253","https://openalex.org/W2127141656","https://openalex.org/W2131342762","https://openalex.org/W2327501763","https://openalex.org/W2514741789","https://openalex.org/W2766219058","https://openalex.org/W2886180730","https://openalex.org/W2889163603","https://openalex.org/W2889282842","https://openalex.org/W2936774411","https://openalex.org/W2962780374","https://openalex.org/W2963747784","https://openalex.org/W2964539095","https://openalex.org/W3015190365","https://openalex.org/W3023126978","https://openalex.org/W3094979069","https://openalex.org/W3095697114","https://openalex.org/W3097643313","https://openalex.org/W3160133086","https://openalex.org/W3160622492","https://openalex.org/W4385245566","https://openalex.org/W6638749077","https://openalex.org/W6739901393","https://openalex.org/W6770425081","https://openalex.org/W6777158299"],"related_works":["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","https://openalex.org/W1482209366","https://openalex.org/W3172695526"],"abstract_inverted_index":{"Recently,":[0],"End-to-End":[1],"(E2E)":[2],"frameworks":[3,58,79],"have":[4],"achieved":[5,132],"remarkable":[6],"results":[7],"on":[8,71,103,123,133],"various":[9,104],"Automatic":[10],"Speech":[11],"Recognition":[12],"(ASR)":[13],"tasks.":[14],"However,":[15],"Lattice-Free":[16],"Maximum":[17],"Mutual":[18],"Information":[19],"(LF-MMI),":[20],"as":[21],"one":[22],"of":[23,73,93,113,119],"the":[24,74,91,94],"discriminative":[25],"training":[26,61],"criteria":[27],"that":[28,90],"show":[29],"superior":[30],"performance":[31,101],"in":[32,39,59],"hybrid":[33],"ASR":[34,41,57,109],"systems,":[35],"is":[36,130,142],"rarely":[37],"adopted":[38],"E2E":[40,56,78,108],"frameworks.":[42,110],"In":[43],"this":[44],"work,":[45],"we":[46],"propose":[47],"a":[48],"novel":[49],"approach":[50,67],"to":[51,99],"introduce":[52],"LF-MMI":[53,95],"criterion":[54,96],"into":[55],"both":[60],"and":[62,84,106,135],"decoding":[63],"stages.":[64],"The":[65,111],"proposed":[66],"shows":[68],"its":[69],"effectiveness":[70],"two":[72],"most":[75],"widely":[76],"used":[77],"including":[80],"Attention-Based":[81],"Encoder-Decoders":[82],"(AEDs)":[83],"Neural":[85],"Transducers":[86],"(NTs).":[87],"Experiments":[88],"suggest":[89],"introduction":[92],"consistently":[97],"leads":[98],"significant":[100,127],"improvements":[102],"datasets":[105,137],"different":[107],"best":[112],"our":[114],"models":[115],"achieves":[116],"competitive":[117],"CER":[118],"4.1%":[120],"/":[121],"4.4%":[122],"Aishell-1":[124],"dev/test":[125],"set;":[126],"error":[128],"reduction":[129],"also":[131],"Aishell-2":[134],"Librispeech":[136],"over":[138],"strong":[139],"baselines.":[140],"Code":[141],"released":[143],"<sup":[144],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[145],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">1</sup>":[146],".":[147]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":5}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}