{"id":"https://openalex.org/W4408353241","doi":"https://doi.org/10.1109/icassp49660.2025.10890510","title":"Preference Alignment Improves Language Model-Based TTS","display_name":"Preference Alignment Improves Language Model-Based TTS","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408353241","doi":"https://doi.org/10.1109/icassp49660.2025.10890510"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10890510","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890510","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","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/A5068192693","display_name":"Jinchuan Tian","orcid":"https://orcid.org/0000-0002-2129-471X"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Jinchuan Tian","raw_affiliation_strings":["Carnegie Mellon University,Language Technologies Institute"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University,Language Technologies Institute","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041339617","display_name":"Chunlei Zhang","orcid":"https://orcid.org/0000-0002-7781-0218"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chunlei Zhang","raw_affiliation_strings":["Tencent AI LAB"],"affiliations":[{"raw_affiliation_string":"Tencent AI LAB","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008336983","display_name":"Jiatong Shi","orcid":"https://orcid.org/0000-0002-9050-8304"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiatong Shi","raw_affiliation_strings":["Carnegie Mellon University,Language Technologies Institute"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University,Language Technologies Institute","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100396872","display_name":"Hao Zhang","orcid":"https://orcid.org/0000-0002-2058-7123"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Zhang","raw_affiliation_strings":["Tencent AI LAB"],"affiliations":[{"raw_affiliation_string":"Tencent AI LAB","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102286496","display_name":"Jianwei Yu","orcid":null},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jianwei Yu","raw_affiliation_strings":["Tencent AI LAB"],"affiliations":[{"raw_affiliation_string":"Tencent AI LAB","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001291873","display_name":"Shinji Watanabe","orcid":"https://orcid.org/0000-0002-5970-8631"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shinji Watanabe","raw_affiliation_strings":["Carnegie Mellon University,Language Technologies Institute"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University,Language Technologies Institute","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5034476404","display_name":"Dong Yu","orcid":"https://orcid.org/0000-0003-0520-6844"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dong Yu","raw_affiliation_strings":["Tencent AI LAB"],"affiliations":[{"raw_affiliation_string":"Tencent AI LAB","institution_ids":["https://openalex.org/I2250653659"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5068192693"],"corresponding_institution_ids":["https://openalex.org/I74973139"],"apc_list":null,"apc_paid":null,"fwci":8.7274,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.9705845,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10215","display_name":"Semantic Web and Ontologies","score":0.6834999918937683,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.6834999918937683,"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":0.6518999934196472,"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":0.7455988526344299},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.4841959774494171},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4541049897670746},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4204499423503876},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4109998345375061},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.0759023129940033},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.06539109349250793}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7455988526344299},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.4841959774494171},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4541049897670746},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4204499423503876},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4109998345375061},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0759023129940033},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.06539109349250793}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10890510","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890510","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":47,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2763110165","https://openalex.org/W2963609956","https://openalex.org/W2972909277","https://openalex.org/W3024869864","https://openalex.org/W3095410713","https://openalex.org/W3175306105","https://openalex.org/W3198694222","https://openalex.org/W3215615641","https://openalex.org/W4225956675","https://openalex.org/W4307323391","https://openalex.org/W4381786045","https://openalex.org/W4390075359","https://openalex.org/W4392903247","https://openalex.org/W4392903389","https://openalex.org/W4392904805","https://openalex.org/W4392908343","https://openalex.org/W4402111462","https://openalex.org/W4402111789","https://openalex.org/W4402111799","https://openalex.org/W4404782689","https://openalex.org/W4406461437","https://openalex.org/W6741002519","https://openalex.org/W6757817989","https://openalex.org/W6763832098","https://openalex.org/W6796464841","https://openalex.org/W6796730497","https://openalex.org/W6805710207","https://openalex.org/W6810738896","https://openalex.org/W6847363464","https://openalex.org/W6848735303","https://openalex.org/W6850334629","https://openalex.org/W6852418670","https://openalex.org/W6853096648","https://openalex.org/W6853188576","https://openalex.org/W6853666897","https://openalex.org/W6853998256","https://openalex.org/W6857273126","https://openalex.org/W6859557930","https://openalex.org/W6860276258","https://openalex.org/W6861548716","https://openalex.org/W6864145738","https://openalex.org/W6864829915","https://openalex.org/W6868500021","https://openalex.org/W6869189190","https://openalex.org/W6869301554","https://openalex.org/W6870266335"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Recent":[0],"advancements":[1],"in":[2,96],"text-to-speech":[3],"(TTS)":[4],"have":[5],"shown":[6],"that":[7,74],"language":[8],"model":[9],"(LM)-based":[10],"systems":[11],"offer":[12],"competitive":[13],"performance":[14],"to":[15,30,106,112],"their":[16],"counterparts.":[17],"Further":[18],"optimization":[19],"can":[20],"be":[21],"achieved":[22],"through":[23],"preference":[24,54,75,102],"alignment":[25,55,76,103],"algorithms,":[26,56],"which":[27],"adjust":[28],"LMs":[29],"align":[31],"with":[32,87],"the":[33,39,42,88],"preferences":[34],"of":[35,41,52],"reward":[36],"models,":[37],"enhancing":[38],"desirability":[40],"generated":[43],"content.":[44],"This":[45],"study":[46],"presents":[47],"a":[48,66],"thorough":[49],"empirical":[50],"evaluation":[51,85],"how":[53],"particularly":[57],"Direct":[58],"Preference":[59],"Optimization":[60],"(DPO),":[61],"enhance":[62],"LM-based":[63,69],"TTS.":[64],"With":[65],"1.15B":[67],"parameter":[68],"TTS":[70],"model,":[71],"we":[72],"demonstrate":[73],"consistently":[77],"improves":[78],"intelligibility,":[79],"speaker":[80],"similarity,":[81],"and":[82,109],"proxy":[83],"subjective":[84],"scores,":[86],"latter":[89],"two":[90],"metrics":[91],"surpassing":[92],"even":[93],"human":[94],"speech":[95],"certain":[97],"evaluations.":[98],"We":[99],"also":[100],"show":[101],"is":[104],"applicable":[105],"low-resource":[107],"scenarios":[108],"effectively":[110],"generalized":[111],"out-of-domain":[113],"applications.":[114]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2}],"updated_date":"2026-04-23T09:07:50.710637","created_date":"2025-10-10T00:00:00"}