{"id":"https://openalex.org/W4412945232","doi":"https://doi.org/10.18653/v1/2025.acl-long.598","title":"Advancing Zero-shot Text-to-Speech Intelligibility across Diverse Domains via Preference Alignment","display_name":"Advancing Zero-shot Text-to-Speech Intelligibility across Diverse Domains via Preference Alignment","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945232","doi":"https://doi.org/10.18653/v1/2025.acl-long.598"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.598","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.598","pdf_url":"https://aclanthology.org/2025.acl-long.598.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":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.598.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100617725","display_name":"Xueyao Zhang","orcid":"https://orcid.org/0000-0002-4495-3853"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xueyao Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090055283","display_name":"Yuancheng Wang","orcid":"https://orcid.org/0000-0002-6253-8311"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuancheng Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005342759","display_name":"Chaoren Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chaoren Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077287145","display_name":"Ziniu Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ziniu Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114269671","display_name":"Zhuo Chen","orcid":"https://orcid.org/0000-0001-8899-2232"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhuo Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5102765381","display_name":"Zhizheng Wu","orcid":"https://orcid.org/0009-0001-1192-9857"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhizheng Wu","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":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"12251","last_page":"12270"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9825000166893005,"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":0.9825000166893005,"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/T12031","display_name":"Speech and dialogue systems","score":0.9599000215530396,"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":0.9498999714851379,"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":0.6114508509635925},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5413065552711487},{"id":"https://openalex.org/keywords/intelligibility","display_name":"Intelligibility (philosophy)","score":0.5140013098716736},{"id":"https://openalex.org/keywords/zero","display_name":"Zero (linguistics)","score":0.47031036019325256},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.4501023292541504},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.37974199652671814},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.17220133543014526},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.09243720769882202},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.09052318334579468}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6114508509635925},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5413065552711487},{"id":"https://openalex.org/C60048801","wikidata":"https://www.wikidata.org/wiki/Q1433889","display_name":"Intelligibility (philosophy)","level":2,"score":0.5140013098716736},{"id":"https://openalex.org/C2780813799","wikidata":"https://www.wikidata.org/wiki/Q3274237","display_name":"Zero (linguistics)","level":2,"score":0.47031036019325256},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.4501023292541504},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37974199652671814},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.17220133543014526},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.09243720769882202},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.09052318334579468},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.598","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.598","pdf_url":"https://aclanthology.org/2025.acl-long.598.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.598","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.598","pdf_url":"https://aclanthology.org/2025.acl-long.598.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":[{"display_name":"Reduced inequalities","score":0.6399999856948853,"id":"https://metadata.un.org/sdg/10"}],"awards":[{"id":"https://openalex.org/G4327894984","display_name":null,"funder_award_id":"62376237","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412945232.pdf","grobid_xml":"https://content.openalex.org/works/W4412945232.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2090624569","https://openalex.org/W3013650182","https://openalex.org/W2989283631","https://openalex.org/W2373674497","https://openalex.org/W2081919107","https://openalex.org/W82312331","https://openalex.org/W4249605382","https://openalex.org/W2768478174","https://openalex.org/W2127461790","https://openalex.org/W2069324367"],"abstract_inverted_index":{"Modern":[0],"zero-shot":[1],"text-to-speech":[2],"(TTS)":[3],"systems,":[4],"despite":[5],"using":[6],"extensive":[7],"pre-training,":[8],"often":[9],"struggle":[10],"in":[11,73],"challenging":[12],"scenarios":[13],"such":[14,109],"as":[15,110],"tongue":[16],"twisters,":[17],"repeated":[18],"words,":[19],"codeswitching,":[20],"and":[21,58,84,113],"cross-lingual":[22],"synthesis,":[23],"leading":[24],"to":[25,44,66,75],"intelligibility":[26],"issues.To":[27],"address":[28],"these":[29],"limitations,":[30],"this":[31],"paper":[32],"leverages":[33],"preference":[34],"alignment":[35,124],"techniques,":[36],"which":[37],"enable":[38],"targeted":[39],"construction":[40],"of":[41,103],"out-of-pretraining-distribution":[42],"data":[43],"enhance":[45],"performance.We":[46],"introduce":[47],"a":[48],"new":[49],"dataset,":[50],"named":[51],"the":[52,60,99,117],"Intelligibility":[53],"Preference":[54,62],"Speech":[55],"Dataset":[56],"(INTP),":[57],"extend":[59],"Direct":[61],"Optimization":[63],"(DPO)":[64],"framework":[65],"accommodate":[67],"diverse":[68,92],"TTS":[69,89],"architectures.After":[70],"INTP":[71,104],"alignment,":[72],"addition":[74],"intelligibility,":[76],"we":[77,96,115],"observe":[78],"overall":[79],"improvements":[80,121],"including":[81],"naturalness,":[82],"similarity,":[83],"audio":[85],"quality":[86],"for":[87,105,119],"multiple":[88],"models":[90,108],"across":[91],"domains.Based":[93],"on":[94,126],"that,":[95],"also":[97],"verify":[98],"weak-to-strong":[100],"generalization":[101],"ability":[102],"more":[106],"intelligible":[107],"CosyVoice":[111],"2":[112],"Ints.Moreover,":[114],"showcase":[116],"potential":[118],"further":[120],"through":[122],"iterative":[123],"based":[125],"Ints.":[127]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}