{"id":"https://openalex.org/W4416035409","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.1600","title":"Towards Language-Agnostic STIPA: Universal Phonetic Transcription to Support Language Documentation at Scale","display_name":"Towards Language-Agnostic STIPA: Universal Phonetic Transcription to Support Language Documentation at Scale","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416035409","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.1600"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.emnlp-main.1600","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1600","pdf_url":"https://aclanthology.org/2025.emnlp-main.1600.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 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.emnlp-main.1600.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120308829","display_name":"Jacob Lee Suchardt","orcid":null},"institutions":[{"id":"https://openalex.org/I881427289","display_name":"University of Gothenburg","ror":"https://ror.org/01tm6cn81","country_code":"SE","type":"education","lineage":["https://openalex.org/I881427289"]}],"countries":["SE"],"is_corresponding":true,"raw_author_name":"Jacob Lee Suchardt","raw_affiliation_strings":["University of Gothenburg"],"affiliations":[{"raw_affiliation_string":"University of Gothenburg","institution_ids":["https://openalex.org/I881427289"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120308830","display_name":"Hana El-Shazli","orcid":null},"institutions":[{"id":"https://openalex.org/I881427289","display_name":"University of Gothenburg","ror":"https://ror.org/01tm6cn81","country_code":"SE","type":"education","lineage":["https://openalex.org/I881427289"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Hana El-Shazli","raw_affiliation_strings":["University of Gothenburg"],"affiliations":[{"raw_affiliation_string":"University of Gothenburg","institution_ids":["https://openalex.org/I881427289"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5080549390","display_name":"Pierluigi Cassotti","orcid":"https://orcid.org/0000-0001-7824-7167"},"institutions":[{"id":"https://openalex.org/I881427289","display_name":"University of Gothenburg","ror":"https://ror.org/01tm6cn81","country_code":"SE","type":"education","lineage":["https://openalex.org/I881427289"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Pierluigi Cassotti","raw_affiliation_strings":["University of Gothenburg"],"affiliations":[{"raw_affiliation_string":"University of Gothenburg","institution_ids":["https://openalex.org/I881427289"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5120308829"],"corresponding_institution_ids":["https://openalex.org/I881427289"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.35431591,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"31411","last_page":"31427"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10730","display_name":"Language Development and Disorders","score":0.38339999318122864,"subfield":{"id":"https://openalex.org/subfields/3204","display_name":"Developmental and Educational Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T10730","display_name":"Language Development and Disorders","score":0.38339999318122864,"subfield":{"id":"https://openalex.org/subfields/3204","display_name":"Developmental and Educational Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10403","display_name":"Phonetics and Phonology Research","score":0.16519999504089355,"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"}},{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.09749999642372131,"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/transcription","display_name":"Transcription (linguistics)","score":0.7429999709129333},{"id":"https://openalex.org/keywords/phonetic-transcription","display_name":"Phonetic transcription","score":0.6715999841690063},{"id":"https://openalex.org/keywords/documentation","display_name":"Documentation","score":0.5819000005722046},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.48249998688697815},{"id":"https://openalex.org/keywords/phonetics","display_name":"Phonetics","score":0.3197000026702881}],"concepts":[{"id":"https://openalex.org/C179926584","wikidata":"https://www.wikidata.org/wiki/Q207714","display_name":"Transcription (linguistics)","level":2,"score":0.7429999709129333},{"id":"https://openalex.org/C2777853878","wikidata":"https://www.wikidata.org/wiki/Q743569","display_name":"Phonetic transcription","level":2,"score":0.6715999841690063},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.616100013256073},{"id":"https://openalex.org/C56666940","wikidata":"https://www.wikidata.org/wiki/Q788790","display_name":"Documentation","level":2,"score":0.5819000005722046},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4869999885559082},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.48249998688697815},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38370001316070557},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.37529999017715454},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3587000072002411},{"id":"https://openalex.org/C137584468","wikidata":"https://www.wikidata.org/wiki/Q35395","display_name":"Phonetics","level":2,"score":0.3197000026702881},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.30630001425743103}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.emnlp-main.1600","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1600","pdf_url":"https://aclanthology.org/2025.emnlp-main.1600.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 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.emnlp-main.1600","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1600","pdf_url":"https://aclanthology.org/2025.emnlp-main.1600.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 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1470719208","display_name":null,"funder_award_id":"M21-0021","funder_id":"https://openalex.org/F4320322659","funder_display_name":"Riksbankens Jubileumsfond"},{"id":"https://openalex.org/G3165808449","display_name":null,"funder_award_id":"022-06725","funder_id":"https://openalex.org/F4320322581","funder_display_name":"Vetenskapsr\u00e5det"},{"id":"https://openalex.org/G4493597635","display_name":null,"funder_award_id":"NAISS","funder_id":"https://openalex.org/F4320322581","funder_display_name":"Vetenskapsr\u00e5det"},{"id":"https://openalex.org/G8338695050","display_name":null,"funder_award_id":"2022-0672","funder_id":"https://openalex.org/F4320322581","funder_display_name":"Vetenskapsr\u00e5det"},{"id":"https://openalex.org/G8869864247","display_name":null,"funder_award_id":"2022-06725","funder_id":"https://openalex.org/F4320322581","funder_display_name":"Vetenskapsr\u00e5det"}],"funders":[{"id":"https://openalex.org/F4320322581","display_name":"Vetenskapsr\u00e5det","ror":"https://ror.org/03zttf063"},{"id":"https://openalex.org/F4320322659","display_name":"Riksbankens Jubileumsfond","ror":"https://ror.org/02jkbm893"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416035409.pdf","grobid_xml":"https://content.openalex.org/works/W4416035409.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"This":[0],"paper":[1],"explores":[2],"the":[3,13,23,112,121],"use":[4,78],"of":[5,15,68,114,123],"existing":[6],"stateof-the-art":[7],"speech":[8,17],"recognition":[9],"models":[10,70,91],"(ASR)":[11],"for":[12,36,48,59,116],"task":[14],"transcribing":[16],"with":[18,97],"narrow":[19],"phonetic":[20,102],"transcriptions":[21,96],"using":[22],"International":[24],"Phonetic":[25],"Alphabet":[26],"(Speech-to-IPA,":[27],"STIPA).Unlike":[28],"conventional":[29],"ASR":[30,90],"systems":[31],"focused":[32],"on":[33,80],"orthographic":[34],"output":[35],"high-resource":[37],"languages,":[38],"STIPA":[39,57,69,115],"can":[40,92],"be":[41],"used":[42],"as":[43],"a":[44,55,65,77,82],"language-agnostic":[45],"interface":[46],"valuable":[47],"documenting":[49],"under-resourced":[50],"and":[51,63,120],"unwritten":[52],"languages.We":[53],"introduce":[54],"new":[56],"dataset":[58],"South":[60],"Levantine":[61],"Arabic":[62],"present":[64],"large-scale":[66],"evaluation":[67],"across":[71],"21":[72],"language":[73,118],"families.Additionally,":[74],"we":[75],"provide":[76],"case":[79],"Sanna,":[81],"severely":[83],"endangered":[84],"language.Our":[85],"findings":[86],"show":[87],"that":[88],"fine-tuned":[89],"produce":[93],"accurate":[94],"IPA":[95],"limited":[98],"supervision,":[99],"significantly":[100],"reducing":[101],"error":[103],"rates":[104],"even":[105],"in":[106],"extremely":[107],"low-resource":[108],"settings.The":[109],"results":[110],"highlight":[111],"potential":[113],"scalable":[117],"documentation":[119],"relevance":[122],"training":[124],"data":[125],"composition.":[126]},"counts_by_year":[],"updated_date":"2026-04-10T15:06:20.359241","created_date":"2025-11-08T00:00:00"}