{"id":"https://openalex.org/W7164037568","doi":"https://doi.org/10.48550/arxiv.2606.09019","title":"TLDR: Compressing Audio Tokens for Efficient Autoregressive Text-to-Speech","display_name":"TLDR: Compressing Audio Tokens for Efficient Autoregressive Text-to-Speech","publication_year":2026,"publication_date":"2026-06-08","ids":{"openalex":"https://openalex.org/W7164037568","doi":"https://doi.org/10.48550/arxiv.2606.09019"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.09019","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.09019","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.09019","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138243503","display_name":"Yejin Lee","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lee, Yejin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138251131","display_name":"Junwon Moon","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Moon, Junwon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138203615","display_name":"Hyoeun Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Hyoeun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129665296","display_name":"Hyunjin Choi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Choi, Hyunjin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138262287","display_name":"Heeseung Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Heeseung","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138259265","display_name":"Kyuhong Shim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shim, Kyuhong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.8037999868392944,"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":0.8037999868392944,"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/T10860","display_name":"Speech and Audio Processing","score":0.13339999318122864,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.00860000029206276,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/autoregressive-model","display_name":"Autoregressive model","score":0.7591000199317932},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.6764000058174133},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5896000266075134},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.5745000243186951},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5263000130653381},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.450300008058548},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.42899999022483826},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4187999963760376}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8052999973297119},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.7591000199317932},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.6764000058174133},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5896000266075134},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5859000086784363},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.5745000243186951},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5263000130653381},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.450300008058548},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.42899999022483826},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4187999963760376},{"id":"https://openalex.org/C161765866","wikidata":"https://www.wikidata.org/wiki/Q184748","display_name":"Codec","level":2,"score":0.41019999980926514},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.4092000126838684},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39570000767707825},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3725000023841858},{"id":"https://openalex.org/C112313634","wikidata":"https://www.wikidata.org/wiki/Q7886648","display_name":"Complement (music)","level":5,"score":0.36570000648498535},{"id":"https://openalex.org/C198082294","wikidata":"https://www.wikidata.org/wiki/Q3399648","display_name":"Position (finance)","level":2,"score":0.3447999954223633},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.3294000029563904},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.32190001010894775},{"id":"https://openalex.org/C35639132","wikidata":"https://www.wikidata.org/wiki/Q7452468","display_name":"Sequence labeling","level":3,"score":0.31439998745918274},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.3107999861240387},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3001999855041504},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.26589998602867126},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.25870001316070557}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.09019","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.09019","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.09019","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.09019","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Codec-based":[0],"autoregressive":[1,70],"(AR)":[2],"speech":[3,14,83,119],"language":[4],"models":[5,101],"have":[6],"achieved":[7],"strong":[8],"text-to-speech":[9],"(TTS)":[10],"quality":[11],"by":[12,76,114,150],"modeling":[13,80,161],"as":[15],"sequences":[16,35,84],"of":[17,132,172],"discrete":[18],"audio":[19],"tokens":[20,92,120],"with":[21,61,107],"large":[22],"pretrained":[23,110,173],"backbones.":[24],"However,":[25],"this":[26],"token-level":[27,82],"formulation":[28],"creates":[29],"a":[30,56,68,98,108,125,129,136,164],"structural":[31],"efficiency":[32],"bottleneck:":[33],"speech-token":[34],"are":[36],"much":[37],"longer":[38],"than":[39],"text":[40],"sequences,":[41],"requiring":[42],"the":[43,62,78,102,141,169,179],"AR":[44],"backbone":[45,112],"to":[46,85,152,167],"perform":[47],"causal":[48,79,160],"computation":[49],"at":[50],"every":[51],"token":[52],"position":[53],"and":[54,116,145],"maintain":[55],"KV":[57],"cache":[58],"that":[59,72,157],"grows":[60],"sequence":[63,106],"length.":[64],"We":[65],"introduce":[66],"TLDR,":[67],"patch-based":[69],"framework":[71],"accelerates":[73],"codec-based":[74,174],"AR-TTS":[75,111,143,175],"shifting":[77],"from":[81],"patch-level":[86,158],"sequences.":[87],"TLDR":[88,134],"groups":[89],"consecutive":[90],"codec":[91],"into":[93],"compact":[94],"latent":[95],"patches":[96],"using":[97,124],"lightweight":[99],"compressor,":[100],"resulting":[103],"shorter":[104],"patch":[105,123,130],"frozen":[109],"adapted":[113],"LoRA,":[115],"reconstructs":[117],"fine-grained":[118],"within":[121],"each":[122],"speaker-conditioned":[126],"extractor.":[127],"With":[128],"size":[131],"4,":[133],"achieves":[135],"1.8x":[137],"inference":[138,170],"speedup":[139],"over":[140],"baseline":[142],"model":[144],"reduces":[146],"global":[147,159],"KV-cache":[148],"memory":[149],"up":[151],"75%.":[153],"Experimental":[154],"results":[155],"indicate":[156],"can":[162],"be":[163],"practical":[165],"way":[166],"reduce":[168],"cost":[171],"systems":[176],"without":[177],"replacing":[178],"existing":[180],"modules.":[181]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-10T00:00:00"}