{"id":"https://openalex.org/W4412889674","doi":"https://doi.org/10.18653/v1/2025.acl-long.1498","title":"Analyzing and Mitigating Inconsistency in Discrete Speech Tokens for Neural Codec Language Models","display_name":"Analyzing and Mitigating Inconsistency in Discrete Speech Tokens for Neural Codec Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412889674","doi":"https://doi.org/10.18653/v1/2025.acl-long.1498"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.1498","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1498","pdf_url":"https://aclanthology.org/2025.acl-long.1498.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.1498.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5081447891","display_name":"Wenrui Liu","orcid":"https://orcid.org/0000-0002-9501-7053"},"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"]},{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Wenrui Liu","raw_affiliation_strings":["Zhejiang University ,  Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University ,  Alibaba Group","institution_ids":["https://openalex.org/I45928872","https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022081273","display_name":"Zhifang Guo","orcid":"https://orcid.org/0000-0002-3728-6319"},"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"]},{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Zhifang Guo","raw_affiliation_strings":["Zhejiang University ,  Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University ,  Alibaba Group","institution_ids":["https://openalex.org/I45928872","https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043896172","display_name":"Jin Xu","orcid":"https://orcid.org/0000-0003-2132-8825"},"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"]},{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Jin Xu","raw_affiliation_strings":["Zhejiang University ,  Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University ,  Alibaba Group","institution_ids":["https://openalex.org/I45928872","https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108604484","display_name":"Yuanjun Lv","orcid":"https://orcid.org/0009-0002-2272-9153"},"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"]},{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Yuanjun Lv","raw_affiliation_strings":["Zhejiang University ,  Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University ,  Alibaba Group","institution_ids":["https://openalex.org/I45928872","https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030042334","display_name":"Yunfei Chu","orcid":"https://orcid.org/0000-0002-3033-2984"},"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"]},{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Yunfei Chu","raw_affiliation_strings":["Zhejiang University ,  Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University ,  Alibaba Group","institution_ids":["https://openalex.org/I45928872","https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101992936","display_name":"Zemin Liu","orcid":"https://orcid.org/0000-0002-6780-3049"},"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"]},{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Zemin Liu","raw_affiliation_strings":["Zhejiang University ,  Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University ,  Alibaba Group","institution_ids":["https://openalex.org/I45928872","https://openalex.org/I4210095624"]}]},{"author_position":"last","author":{"id":null,"display_name":"Junyang Lin","orcid":null},"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"]},{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Junyang Lin","raw_affiliation_strings":["Zhejiang University ,  Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University ,  Alibaba Group","institution_ids":["https://openalex.org/I45928872","https://openalex.org/I4210095624"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":6.8026,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":{"value":0.96542143,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"31035","last_page":"31046"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9650999903678894,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9650999903678894,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9642000198364258,"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/T10028","display_name":"Topic Modeling","score":0.9452999830245972,"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":0.791224479675293},{"id":"https://openalex.org/keywords/codec","display_name":"Codec","score":0.5994244813919067},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5772507786750793},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4740849733352661},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4726390242576599},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3842388391494751},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.080717533826828}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.791224479675293},{"id":"https://openalex.org/C161765866","wikidata":"https://www.wikidata.org/wiki/Q184748","display_name":"Codec","level":2,"score":0.5994244813919067},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5772507786750793},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4740849733352661},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4726390242576599},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3842388391494751},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.080717533826828}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.1498","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1498","pdf_url":"https://aclanthology.org/2025.acl-long.1498.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.1498","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1498","pdf_url":"https://aclanthology.org/2025.acl-long.1498.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412889674.pdf","grobid_xml":"https://content.openalex.org/works/W4412889674.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2964213236","https://openalex.org/W2163719598","https://openalex.org/W3161919736","https://openalex.org/W2387018512","https://openalex.org/W2107680156","https://openalex.org/W4301184752","https://openalex.org/W2288771647","https://openalex.org/W2751422192","https://openalex.org/W1509797384","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Building":[0],"upon":[1],"advancements":[2],"in":[3,17,36,95,102],"Large":[4],"Language":[5],"Models":[6],"(LLMs),":[7],"the":[8,111,124,128,135,149],"field":[9],"of":[10,127,153,157],"audio":[11,32,67,116,130,158],"processing":[12],"has":[13],"seen":[14],"increased":[15],"interest":[16],"training":[18],"speech":[19,24,29,52,84],"generation":[20],"tasks":[21],"with":[22],"discrete":[23,51],"token":[25,47],"sequences.However,":[26],"directly":[27],"discretizing":[28],"by":[30,88],"neural":[31,96,129,136],"codecs":[33],"often":[34],"results":[35,101],"sequences":[37,48],"that":[38],"fundamentally":[39],"differ":[40],"from":[41],"text":[42,46],"sequences.Unlike":[43],"text,":[44],"where":[45],"are":[49],"deterministic,":[50],"tokens":[53],"can":[54,79],"exhibit":[55],"significant":[56],"variability":[57],"based":[58],"on":[59,134],"contextual":[60],"factors,":[61],"while":[62],"still":[63],"producing":[64],"perceptually":[65],"identical":[66],"segments.We":[68],"refer":[69],"to":[70,81],"this":[71,106],"phenomenon":[72,113,126],"as":[73,119],"Discrete":[74],"Representation":[75],"Inconsistency":[76],"(DRI).This":[77],"inconsistency":[78],"lead":[80],"a":[82],"single":[83],"segment":[85],"being":[86],"represented":[87],"multiple":[89],"divergent":[90],"sequences,":[91],"which":[92],"creates":[93],"confusion":[94],"codec":[97,137],"language":[98,138],"models":[99],"and":[100,142,151],"poor":[103],"generated":[104],"speech.In":[105],"paper,":[107],"we":[108],"quantitatively":[109],"analyze":[110],"DRI":[112,125],"within":[114],"popular":[115],"tokenizers":[117],"such":[118],"En-Codec.Our":[120],"approach":[121],"effectively":[122],"mitigates":[123],"codec.Furthermore,":[131],"extensive":[132],"experiments":[133],"model":[139],"over":[140],"LibriTTS":[141],"large-scale":[143],"MLS":[144],"dataset":[145],"(44,000":[146],"hours)":[147],"demonstrate":[148],"effectiveness":[150],"generality":[152],"our":[154],"method.The":[155],"demo":[156],"samples":[159],"is":[160],"available":[161],"online":[162],"1":[163],".":[164]},"counts_by_year":[{"year":2026,"cited_by_count":4}],"updated_date":"2026-06-18T10:00:31.954636","created_date":"2025-10-10T00:00:00"}