{"id":"https://openalex.org/W7224419786","doi":"https://doi.org/10.18653/2026.winlp-main.7","title":"AmhTTS: A Low-Resource Amharic Text-to-Speech System","display_name":"AmhTTS: A Low-Resource Amharic Text-to-Speech System","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7224419786","doi":"https://doi.org/10.18653/2026.winlp-main.7"},"language":"en","primary_location":{"id":"doi:10.18653/2026.winlp-main.7","doi":"https://doi.org/10.18653/2026.winlp-main.7","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.winlp-main.7","pdf_url":"https://aclanthology.org/2026.winlp-main.7.pdf","source":{"id":"https://openalex.org/S7407114509","display_name":"Widening NLP Workshop (WiNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 10th Widening NLP Workshop","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://aclanthology.org/2026.winlp-main.7.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5011399695","display_name":"Walelign Tewabe Sewunetie","orcid":"https://orcid.org/0000-0003-2787-8019","observed_orcids":["https://orcid.org/0000-0003-2787-8019"]},"institutions":[{"id":"https://openalex.org/I4210112059","display_name":"African Institute for Mathematical Sciences","ror":"https://ror.org/02w32z542","country_code":"RW","type":"education","lineage":["https://openalex.org/I4210112059"]},{"id":"https://openalex.org/I4405261291","display_name":"AIMS Research and Innovation Centre","ror":"https://ror.org/052gr0y82","country_code":"RW","type":"education","lineage":["https://openalex.org/I4405261291"]}],"countries":["RW"],"is_corresponding":false,"raw_author_name":"Walelign Tewabe Sewunetie","raw_affiliation_strings":["AIMS Research & Innovation Centre , Kigali , Rwanda"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AIMS Research & Innovation Centre , Kigali , Rwanda","institution_ids":["https://openalex.org/I4210112059","https://openalex.org/I4405261291"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012797755","display_name":"Rahel Mekonen Tamiru","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rahel Mekonen Tamiru","raw_affiliation_strings":["Ethiopian Artificial Intelligence Institute (EAII) , Addis Abeba , Ethiopia","Meta Codec Avatars Lab in Pittsburgh , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ethiopian Artificial Intelligence Institute (EAII) , Addis Abeba , Ethiopia","institution_ids":[]},{"raw_affiliation_string":"Meta Codec Avatars Lab in Pittsburgh , USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5072548521","display_name":"Israel D. Gebru","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Israel D. Gebru","raw_affiliation_strings":["Ethiopian Artificial Intelligence Institute (EAII) , Addis Abeba , Ethiopia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ethiopian Artificial Intelligence Institute (EAII) , Addis Abeba , Ethiopia","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.89943777,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"68","last_page":"76"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":1.0,"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":1.0,"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":0.0,"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/T10403","display_name":"Phonetics and Phonology Research","score":0.0,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/text-to-speech","display_name":"text-to-speech","score":0.9169999957084656},{"id":"https://openalex.org/keywords/amharic","display_name":"Amharic language","score":0.8769999742507935},{"id":"https://openalex.org/keywords/low-resource-languages","display_name":"low-resource languages","score":0.9610000252723694},{"id":"https://openalex.org/keywords/speech-synthesis","display_name":"speech synthesis","score":0.9039999842643738}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.37470000982284546},{"id":"https://openalex.org/C2780900699","wikidata":"https://www.wikidata.org/wiki/Q28244","display_name":"Amharic","level":2,"score":0.2980000078678131},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.2888000011444092},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28119999170303345},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2736999988555908},{"id":"https://openalex.org/C127313418","wikidata":"https://www.wikidata.org/wiki/Q1069","display_name":"Geology","level":0,"score":0.24420000612735748},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.22010000050067902},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.2190999984741211},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.21899999678134918},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.20919999480247498}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/2026.winlp-main.7","doi":"https://doi.org/10.18653/2026.winlp-main.7","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.winlp-main.7","pdf_url":"https://aclanthology.org/2026.winlp-main.7.pdf","source":{"id":"https://openalex.org/S7407114509","display_name":"Widening NLP Workshop (WiNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 10th Widening NLP Workshop","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/2026.winlp-main.7","doi":"https://doi.org/10.18653/2026.winlp-main.7","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.winlp-main.7","pdf_url":"https://aclanthology.org/2026.winlp-main.7.pdf","source":{"id":"https://openalex.org/S7407114509","display_name":"Widening NLP Workshop (WiNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 10th Widening NLP Workshop","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7224419786.pdf","grobid_xml":"https://content.openalex.org/works/W7224419786.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2061706163","https://openalex.org/W2564639338","https://openalex.org/W71670877","https://openalex.org/W7103994941","https://openalex.org/W7206209981","https://openalex.org/W789263950","https://openalex.org/W4309739688","https://openalex.org/W2138559503","https://openalex.org/W2972745048","https://openalex.org/W2396764420"],"abstract_inverted_index":{"The":[0],"scarcity":[1],"of":[2,90,160,165,177,189,193],"high-quality,":[3],"phonetically":[4],"diverse":[5],"datasets":[6],"for":[7,217,224],"low-resource":[8,183,242],"languages":[9],"like":[10],"Amharic":[11,27,39,53,83,108],"remains":[12],"a":[13,35,43,96,115,120,168,182,191],"significant":[14],"barrier":[15],"to":[16,77,181,240],"high-fidelity":[17],"speech":[18,84],"synthesis.We":[19],"present":[20],"AmhTTS,":[21],"the":[22,48,74,87,91,173,194,199,203,220,234],"first":[23,75],"open,":[24],"natively":[25,68],"evaluated":[26],"text-to-speech":[28],"system,":[29,55],"developed":[30],"by":[31],"fine-tuning":[32],"SpeechT5":[33,178],"on":[34,69,82,167],"curated":[36],"25-hour,":[37],"single-speaker":[38],"corpus":[40],"paired":[41],"with":[42,185],"HiFi-GAN":[44],"vocoder.Unlike":[45],"Meta":[46],"MMS,":[47],"only":[49,186],"other":[50],"publicly":[51,228],"available":[52],"TTS":[54,243],"which":[56],"requires":[57],"Latin":[58],"script":[59,71],"transliteration":[60],"and":[61,72,107,142,146,149,152,162,212,237],"reports":[62],"no":[63],"Amharic-specific":[64],"evaluation,":[65],"AmhTTS":[66,235],"operates":[67],"Ethiopic":[70,92],"is":[73,112,247],"system":[76,204],"report":[78],"human":[79,117],"MOS":[80,159,164],"scores":[81],"synthesis.To":[85],"address":[86],"orthographic":[88],"complexities":[89],"script,":[93],"we":[94,227],"implement":[95],"threestage":[97],"text":[98],"normalisation":[99],"pipeline":[100],"covering":[101],"homophone":[102],"unification,":[103],"Ge'ez":[104],"numeral":[105,148],"expansion,":[106],"punctuation":[109],"handling.System":[110],"quality":[111],"assessed":[113],"through":[114],"rigorous":[116],"evaluation":[118,238],"via":[119],"purpose-built":[121],"web":[122,245],"platform,":[123],"involving":[124],"28":[125],"native":[126],"Amharic-speaking":[127],"voluntary":[128],"evaluators":[129],"who":[130],"rated":[131],"35":[132],"linguistically":[133],"motivated":[134],"samples":[135],"across":[136],"four":[137],"domains:":[138],"phonetic":[139,207],"coverage":[140],"(ejectives":[141],"labialized":[143],"consonants),":[144],"gemination":[145,213],"prosody,":[147],"date":[150],"reading,":[151],"conversational":[153],"fluency.Results":[154],"yield":[155],"an":[156],"overall":[157],"naturalness":[158],"3.71":[161],"intelligibility":[163],"4.02":[166],"5point":[169],"scale,":[170],"demonstrating":[171],"that":[172],"rich":[174],"pretrained":[175],"representations":[176],"transfer":[179],"effectively":[180],"language":[184],"25":[187],"hours":[188,196],"data":[190,239],"fraction":[192],"460":[195],"used":[197],"in":[198,257],"original":[200],"English":[201],"fine-tuning.While":[202],"achieves":[205],"high":[206],"accuracy":[208],"(MOS:":[209],"4.22/4.36),numeral":[210],"synthesis":[211],"modeling":[214],"remain":[215],"areas":[216],"future":[218],"improvement.Given":[219],"demonstrated":[221],"community":[222],"demand":[223],"accessibility":[225],"tools,":[226],"release":[229],"our":[230],"fine-tuned":[231],"model":[232],"weights,":[233],"corpus,":[236],"advance":[241],"research.Our":[244],"application":[246],"currently":[248],"active,":[249],"serving":[250],"users":[251],"at":[252],"multiple":[253],"university":[254],"disability":[255],"centers":[256],"Ethiopia.https://huggingface.co/amhtts.":[258]},"counts_by_year":[],"updated_date":"2026-11-01T11:40:01.658920","created_date":"2026-10-30T00:00:00"}