{"id":"https://openalex.org/W7126414086","doi":"https://doi.org/10.18653/v1/2024.findings-eacl.6","title":"ParrotTTS: Text-to-speech synthesis exploiting disentangled self-supervised representations","display_name":"ParrotTTS: Text-to-speech synthesis exploiting disentangled self-supervised representations","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W7126414086","doi":"https://doi.org/10.18653/v1/2024.findings-eacl.6"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2024.findings-eacl.6","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.findings-eacl.6","pdf_url":"https://aclanthology.org/2024.findings-eacl.6.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":"Findings of the Association for Computational Linguistics: EACL 2024","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2024.findings-eacl.6.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5124672688","display_name":"Neil Shah","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Neil Shah","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124595829","display_name":"Saiteja Kosgi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Saiteja Kosgi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088806439","display_name":"Vishal Tambrahalli","orcid":"https://orcid.org/0009-0008-1396-1106"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vishal Tambrahalli","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124476665","display_name":"Neha S","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Neha S","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020172237","display_name":"Anil Nelakanti","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Anil Nelakanti","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5067298540","display_name":"Vineet Gandhi","orcid":"https://orcid.org/0000-0001-8861-7731"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vineet Gandhi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.3055,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.70261936,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"79","last_page":"91"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.23360000550746918,"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":0.23360000550746918,"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":0.1800999939441681,"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":0.14650000631809235,"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/representation","display_name":"Representation (politics)","score":0.3116999864578247},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.2930999994277954},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.26980000734329224},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.26440000534057617}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.553600013256073},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3709999918937683},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.33160001039505005},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.3116999864578247},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2939999997615814},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.2930999994277954},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.26980000734329224},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.26499998569488525},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.26440000534057617},{"id":"https://openalex.org/C90673727","wikidata":"https://www.wikidata.org/wiki/Q901718","display_name":"Product (mathematics)","level":2,"score":0.24979999661445618}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2024.findings-eacl.6","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.findings-eacl.6","pdf_url":"https://aclanthology.org/2024.findings-eacl.6.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":"Findings of the Association for Computational Linguistics: EACL 2024","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2024.findings-eacl.6","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.findings-eacl.6","pdf_url":"https://aclanthology.org/2024.findings-eacl.6.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":"Findings of the Association for Computational Linguistics: EACL 2024","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","score":0.7511491775512695,"display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7126414086.pdf","grobid_xml":"https://content.openalex.org/works/W7126414086.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0],"present":[1,75],"ParrotTTS,":[2],"a":[3,15,22,27,69,91],"modularized":[4],"text-tospeech":[5],"synthesis":[6],"model":[7],"leveraging":[8],"disentangled":[9],"self-supervised":[10,43],"speech":[11,67],"representations.It":[12],"can":[13,53,103],"train":[14],"multi-speaker":[16],"variant":[17],"effectively":[18],"using":[19,68,89],"transcripts":[20],"from":[21,99],"single":[23],"speaker.Par-rotTTS":[24],"adapts":[25],"to":[26,36],"new":[28],"language":[29],"in":[30,78],"low":[31],"resource":[32],"setup":[33],"and":[34,73,80,101],"generalizes":[35],"languages":[37,57],"not":[38],"seen":[39],"while":[40,58],"training":[41,46],"the":[42,60],"backbone.Moreover,":[44],"without":[45],"on":[47],"bilingual":[48],"or":[49],"parallel":[50],"examples,":[51],"ParrotTTS":[52,100],"transfer":[54],"voices":[55],"across":[56],"preserving":[59],"speaker-specific":[61],"characteristics,":[62],"e.g.,":[63],"synthesizing":[64],"fluent":[65],"Hindi":[66],"French":[70],"speaker's":[71],"voice":[72],"accent.We":[74],"extensive":[76],"results":[77],"monolingual":[79],"multi-lingual":[81,85],"scenarios.ParrotTTS":[82],"outperforms":[83],"state-of-theart":[84],"text-to-speech":[86],"(TTS)":[87],"models":[88],"only":[90],"fraction":[92],"of":[93],"paired":[94],"data":[95],"as":[96],"latter.Speech":[97],"samples":[98],"code":[102],"be":[104],"found":[105],"at":[106],"https://parrot-tts.":[107],"github.io/tts/":[108]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-02-02T00:00:00"}