{"id":"https://openalex.org/W7165141801","doi":"https://doi.org/10.48550/arxiv.2606.18485","title":"MagpieTTS-LF: Inference-Time Long-Form Speech Generation Without Training on Long-Form data","display_name":"MagpieTTS-LF: Inference-Time Long-Form Speech Generation Without Training on Long-Form data","publication_year":2026,"publication_date":"2026-06-16","ids":{"openalex":"https://openalex.org/W7165141801","doi":"https://doi.org/10.48550/arxiv.2606.18485"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.18485","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.18485","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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.2606.18485","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5082093509","display_name":"S Ghosh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ghosh, Subhankar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138926115","display_name":"Jason Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Jason","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030710635","display_name":"Paarth Neekhara","orcid":"https://orcid.org/0009-0002-8598-0353"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Neekhara, Paarth","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015232966","display_name":"Shehzeen Hussain","orcid":"https://orcid.org/0000-0003-4693-2113"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hussain, Shehzeen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040023052","display_name":"Ryan Langman","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Langman, Ryan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030018760","display_name":"Xiaomin Yang","orcid":"https://orcid.org/0000-0002-5593-5108"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Xuesong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5054924576","display_name":"Roy Fejgin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fejgin, Roy","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":0.5949000120162964,"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":0.5949000120162964,"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":0.16899999976158142,"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":0.06419999897480011,"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/naturalness","display_name":"Naturalness","score":0.7131999731063843},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5713000297546387},{"id":"https://openalex.org/keywords/sentence","display_name":"Sentence","score":0.4952000081539154},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.43529999256134033},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4050999879837036},{"id":"https://openalex.org/keywords/prosody","display_name":"Prosody","score":0.3952000141143799},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.3529999852180481},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.33559998869895935}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7871000170707703},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.7360000014305115},{"id":"https://openalex.org/C134537474","wikidata":"https://www.wikidata.org/wiki/Q17144832","display_name":"Naturalness","level":2,"score":0.7131999731063843},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5713000297546387},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.4952000081539154},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4799000024795532},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.474700003862381},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.43529999256134033},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4050999879837036},{"id":"https://openalex.org/C542774811","wikidata":"https://www.wikidata.org/wiki/Q10880526","display_name":"Prosody","level":2,"score":0.3952000141143799},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.3529999852180481},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.33559998869895935},{"id":"https://openalex.org/C14999030","wikidata":"https://www.wikidata.org/wiki/Q16346","display_name":"Speech synthesis","level":2,"score":0.3319000005722046},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3228999972343445},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.3140000104904175},{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.2971000075340271},{"id":"https://openalex.org/C22927095","wikidata":"https://www.wikidata.org/wiki/Q1784206","display_name":"Stateful firewall","level":3,"score":0.2919999957084656},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.2847999930381775},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.2833000123500824},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.27090001106262207},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.26499998569488525},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.2517000138759613}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.18485","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.18485","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.18485","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.18485","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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":{"Neural":[0],"Text-to-Speech":[1],"(TTS)":[2],"systems":[3],"achieve":[4],"remarkable":[5],"quality":[6],"on":[7,102],"short":[8],"utterances":[9],"but":[10],"long-form":[11,50],"speech":[12,51],"generation":[13],"shows":[14],"prosodic":[15,87,99,111],"drift,":[16],"speaker":[17,113],"inconsistencies":[18],"and":[19,72,115],"sentence":[20,84],"boundary":[21,116],"artifacts.":[22],"Existing":[23],"approaches":[24],"either":[25],"compress":[26],"sequences,":[27],"increase":[28],"context":[29,82],"length":[30],"or":[31],"naively":[32],"concatenate":[33],"independently":[34],"synthesized":[35],"chunks.":[36],"We":[37],"present":[38],"an":[39],"inference-time":[40],"approach":[41],"called":[42],"MagpieTTS-LF":[43],"that":[44,80,93],"enables":[45],"MagpieTTS":[46],"to":[47,65,119],"produce":[48],"coherent":[49],"without":[52],"model":[53],"retraining.":[54],"Our":[55],"method":[56],"introduces":[57],"three":[58],"key":[59],"innovations:":[60],"(1)":[61],"soft":[62],"attention":[63],"priors":[64],"guide":[66],"monotonic":[67],"alignment":[68],"while":[69],"preserving":[70],"past":[71,95],"future":[73],"context;":[74],"(2)":[75],"a":[76],"stateful":[77],"inference":[78],"algorithm":[79],"maintains":[81],"across":[83],"chunks,":[85],"ensuring":[86],"continuity;":[88],"(3)":[89],"history-aware":[90],"text":[91,96],"encoding":[92],"uses":[94],"for":[97],"discourse-level":[98],"planning.":[100],"Experiments":[101],"long":[103],"texts":[104],"show":[105],"significant":[106],"improvements":[107],"in":[108],"long-range":[109],"intelligibility,":[110],"coherence,":[112],"consistency,":[114],"naturalness":[117],"compared":[118],"other":[120],"baselines.":[121]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-19T00:00:00"}