{"id":"https://openalex.org/W4392902656","doi":"https://doi.org/10.1109/icassp48485.2024.10447296","title":"Loss Masking Is Not Needed In Decoder-Only Transformer For Discrete-Token-Based ASR","display_name":"Loss Masking Is Not Needed In Decoder-Only Transformer For Discrete-Token-Based ASR","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392902656","doi":"https://doi.org/10.1109/icassp48485.2024.10447296"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10447296","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447296","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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/A5100428456","display_name":"Qian Chen","orcid":"https://orcid.org/0000-0002-1909-302X"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qian Chen","raw_affiliation_strings":["Alibaba Group,Speech Lab","Speech Lab, Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group,Speech Lab","institution_ids":["https://openalex.org/I4210095624"]},{"raw_affiliation_string":"Speech Lab, Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100952183","display_name":"Wen Wang","orcid":"https://orcid.org/0000-0002-6545-2265"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wen Wang","raw_affiliation_strings":["Alibaba Group,Speech Lab","Speech Lab, Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group,Speech Lab","institution_ids":["https://openalex.org/I4210095624"]},{"raw_affiliation_string":"Speech Lab, Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100672350","display_name":"Qinglin Zhang","orcid":"https://orcid.org/0000-0002-3640-2883"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qinglin Zhang","raw_affiliation_strings":["Alibaba Group,Speech Lab","Speech Lab, Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group,Speech Lab","institution_ids":["https://openalex.org/I4210095624"]},{"raw_affiliation_string":"Speech Lab, Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014586639","display_name":"Siqi Zheng","orcid":"https://orcid.org/0000-0002-4467-8505"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Siqi Zheng","raw_affiliation_strings":["Alibaba Group,Speech Lab","Speech Lab, Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group,Speech Lab","institution_ids":["https://openalex.org/I4210095624"]},{"raw_affiliation_string":"Speech Lab, Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055433405","display_name":"Shiliang Zhang","orcid":"https://orcid.org/0000-0001-9053-9314"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shiliang Zhang","raw_affiliation_strings":["Alibaba Group,Speech Lab","Speech Lab, Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group,Speech Lab","institution_ids":["https://openalex.org/I4210095624"]},{"raw_affiliation_string":"Speech Lab, Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100686893","display_name":"Chong Deng","orcid":"https://orcid.org/0000-0002-7287-7042"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chong Deng","raw_affiliation_strings":["Alibaba Group,Speech Lab","Speech Lab, Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group,Speech Lab","institution_ids":["https://openalex.org/I4210095624"]},{"raw_affiliation_string":"Speech Lab, Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101568886","display_name":"Yukun Ma","orcid":"https://orcid.org/0000-0002-4419-4287"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yukun Ma","raw_affiliation_strings":["Alibaba Group,Speech Lab","Speech Lab, Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group,Speech Lab","institution_ids":["https://openalex.org/I4210095624"]},{"raw_affiliation_string":"Speech Lab, Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080895253","display_name":"Hai Yu","orcid":"https://orcid.org/0000-0002-8024-1781"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hai Yu","raw_affiliation_strings":["Alibaba Group,Speech Lab","Speech Lab, Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group,Speech Lab","institution_ids":["https://openalex.org/I4210095624"]},{"raw_affiliation_string":"Speech Lab, Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101949746","display_name":"Jiaqing Liu","orcid":"https://orcid.org/0000-0001-8600-8932"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiaqing Liu","raw_affiliation_strings":["Alibaba Group,Speech Lab","Speech Lab, Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group,Speech Lab","institution_ids":["https://openalex.org/I4210095624"]},{"raw_affiliation_string":"Speech Lab, Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100393899","display_name":"Chong Zhang","orcid":"https://orcid.org/0000-0002-2162-4344"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chong Zhang","raw_affiliation_strings":["Alibaba Group,Speech Lab","Speech Lab, Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group,Speech Lab","institution_ids":["https://openalex.org/I4210095624"]},{"raw_affiliation_string":"Speech Lab, Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210095624"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11056","last_page":"11060"},"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":0.9991999864578247,"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":0.9987999796867371,"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":0.7920982241630554},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.7424077987670898},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6632238626480103},{"id":"https://openalex.org/keywords/information-loss","display_name":"Information loss","score":0.5400868058204651},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5262899994850159},{"id":"https://openalex.org/keywords/cross-entropy","display_name":"Cross entropy","score":0.5193013548851013},{"id":"https://openalex.org/keywords/discretization","display_name":"Discretization","score":0.5095646381378174},{"id":"https://openalex.org/keywords/speech-coding","display_name":"Speech coding","score":0.44041121006011963},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38185328245162964},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.2502528429031372}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7920982241630554},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.7424077987670898},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6632238626480103},{"id":"https://openalex.org/C2988416141","wikidata":"https://www.wikidata.org/wiki/Q6031139","display_name":"Information loss","level":2,"score":0.5400868058204651},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5262899994850159},{"id":"https://openalex.org/C167981619","wikidata":"https://www.wikidata.org/wiki/Q1685498","display_name":"Cross entropy","level":3,"score":0.5193013548851013},{"id":"https://openalex.org/C73000952","wikidata":"https://www.wikidata.org/wiki/Q17007827","display_name":"Discretization","level":2,"score":0.5095646381378174},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.44041121006011963},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38185328245162964},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2502528429031372},{"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},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10447296","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447296","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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":31,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1821462560","https://openalex.org/W2113158412","https://openalex.org/W2183341477","https://openalex.org/W2407080277","https://openalex.org/W2747329762","https://openalex.org/W2962784628","https://openalex.org/W2963979492","https://openalex.org/W2988736778","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4307323391","https://openalex.org/W4319862255","https://openalex.org/W4322718191","https://openalex.org/W4323066695","https://openalex.org/W4378501656","https://openalex.org/W4381827575","https://openalex.org/W4385245566","https://openalex.org/W4385822683","https://openalex.org/W4387595589","https://openalex.org/W4389524500","https://openalex.org/W6638523607","https://openalex.org/W6739901393","https://openalex.org/W6769196770","https://openalex.org/W6770514103","https://openalex.org/W6850218400","https://openalex.org/W6850625674","https://openalex.org/W6852800892","https://openalex.org/W6853611000","https://openalex.org/W6853998256","https://openalex.org/W6857968694"],"related_works":["https://openalex.org/W2951959408","https://openalex.org/W2895831313","https://openalex.org/W4289406078","https://openalex.org/W4386453465","https://openalex.org/W3194440789","https://openalex.org/W4390839675","https://openalex.org/W3163951405","https://openalex.org/W4287659163","https://openalex.org/W4226279124","https://openalex.org/W4288264391"],"abstract_inverted_index":{"Recently,":[0],"unified":[1],"speech-text":[2],"models,":[3],"such":[4],"as":[5],"SpeechGPT,":[6],"VioLA,":[7],"and":[8,27,35,146],"AudioPaLM,":[9],"have":[10],"achieved":[11],"remarkable":[12],"performance":[13,105],"on":[14,45,55,95,134],"various":[15],"speech":[16,21,36,49,69,78,97,135,144,158],"tasks.":[17,50],"These":[18],"models":[19,53,143],"discretize":[20],"signals":[22],"into":[23],"tokens":[24,79,98,145],"(speech":[25],"discretization)":[26],"use":[28],"a":[29,41,46,117,127],"shared":[30],"vocabulary":[31],"for":[32,60,150],"both":[33],"text":[34],"tokens.":[37,70,136],"Then":[38],"they":[39],"train":[40],"single":[42],"decoder-only":[43,151],"Transformer":[44],"mixture":[47],"of":[48],"However,":[51],"these":[52],"rely":[54],"the":[56,61,66,91,103,107],"Loss":[57,108,148],"Masking":[58,109,149],"strategy":[59],"ASR":[62,104,154],"task,":[63],"which":[64,125],"ignores":[65],"dependency":[67],"among":[68],"In":[71],"this":[72,113],"paper,":[73],"we":[74,115],"propose":[75,116],"to":[76,85],"model":[77],"in":[80,153],"an":[81],"autoregressive":[82],"way,":[83],"similar":[84],"text.":[86],"We":[87],"find":[88],"that":[89,140],"applying":[90],"conventional":[92],"cross-entropy":[93],"loss":[94,130],"input":[96],"does":[99],"not":[100],"consistently":[101],"improve":[102],"over":[106],"approach.":[110],"To":[111],"address":[112],"issue,":[114],"novel":[118],"approach":[119],"denoted":[120],"Smoothed":[121],"Label":[122],"Distillation":[123],"(SLD),":[124],"applies":[126],"KL":[128],"divergence":[129],"with":[131,156],"smoothed":[132],"labels":[133],"Our":[137],"experiments":[138],"show":[139],"SLD":[141],"effectively":[142],"outperforms":[147],"Transformers":[152],"tasks":[155],"different":[157],"discretization":[159],"methods":[160],"<sup":[161],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[162],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">1</sup>":[163],".":[164]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2024,"cited_by_count":4}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}