{"id":"https://openalex.org/W2405860391","doi":"https://doi.org/10.21437/interspeech.2012-657","title":"Comparison of grapheme-to-phoneme methods on large pronunciation dictionaries and LVCSR tasks","display_name":"Comparison of grapheme-to-phoneme methods on large pronunciation dictionaries and LVCSR tasks","publication_year":2012,"publication_date":"2012-09-09","ids":{"openalex":"https://openalex.org/W2405860391","doi":"https://doi.org/10.21437/interspeech.2012-657","mag":"2405860391"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2012-657","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2012-657","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2012","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/A5020202476","display_name":"Stefan Hahn","orcid":"https://orcid.org/0000-0002-3437-3951"},"institutions":[{"id":"https://openalex.org/I887968799","display_name":"RWTH Aachen University","ror":"https://ror.org/04xfq0f34","country_code":"DE","type":"education","lineage":["https://openalex.org/I887968799"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Stefan Hahn","raw_affiliation_strings":["RWTH Aachen University"],"affiliations":[{"raw_affiliation_string":"RWTH Aachen University","institution_ids":["https://openalex.org/I887968799"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003960312","display_name":"Paul Vozila","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Paul Vozila","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5082506063","display_name":"M. Bisani","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Maximilian Bisani","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5020202476"],"corresponding_institution_ids":["https://openalex.org/I887968799"],"apc_list":null,"apc_paid":null,"fwci":5.3055,"has_fulltext":false,"cited_by_count":28,"citation_normalized_percentile":{"value":0.95657023,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"2538","last_page":"2541"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9997000098228455,"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":0.9997000098228455,"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":0.998199999332428,"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":0.9959999918937683,"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/grapheme","display_name":"Grapheme","score":0.9594016671180725},{"id":"https://openalex.org/keywords/pronunciation","display_name":"Pronunciation","score":0.9023441076278687},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8234038352966309},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5465450286865234},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5417202115058899},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5128741264343262},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.1679648756980896}],"concepts":[{"id":"https://openalex.org/C2776779415","wikidata":"https://www.wikidata.org/wiki/Q2545446","display_name":"Grapheme","level":3,"score":0.9594016671180725},{"id":"https://openalex.org/C2780844864","wikidata":"https://www.wikidata.org/wiki/Q184377","display_name":"Pronunciation","level":2,"score":0.9023441076278687},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8234038352966309},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5465450286865234},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5417202115058899},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5128741264343262},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.1679648756980896},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C30080830","wikidata":"https://www.wikidata.org/wiki/Q169917","display_name":"Graphene","level":2,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.21437/interspeech.2012-657","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2012-657","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2012","raw_type":"proceedings-article"},{"id":"pmh:oai:CiteSeerX.psu:10.1.1.248.3895","is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.248.3895","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"http://www-i6.informatik.rwth-aachen.de/publications/download/811/Hahn--2012.pdf","raw_type":"text"},{"id":"pmh:oai:publications.rwth-aachen.de:207155","is_oa":false,"landing_page_url":"https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-200646%22","pdf_url":null,"source":{"id":"https://openalex.org/S4306401362","display_name":"RWTH Publications (RWTH Aachen)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I887968799","host_organization_name":"RWTH Aachen University","host_organization_lineage":["https://openalex.org/I887968799"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"13th Annual Conference of the International Speech Communication Association, September 9-13, 2012, Portland, Oregon<br/>13. Annual Conference of the International Speech Communication Association, Portland, OR, USA, 2012-09-09 - 2012-09-13","raw_type":"info:eu-repo/semantics/publishedVersion"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.800000011920929}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":12,"referenced_works":["https://openalex.org/W44395966","https://openalex.org/W67332896","https://openalex.org/W72985070","https://openalex.org/W101576916","https://openalex.org/W172866779","https://openalex.org/W201187342","https://openalex.org/W1582482241","https://openalex.org/W1904457459","https://openalex.org/W2090755665","https://openalex.org/W2118947254","https://openalex.org/W2139671364","https://openalex.org/W2150859660"],"related_works":["https://openalex.org/W2506515307","https://openalex.org/W4385893898","https://openalex.org/W2060656088","https://openalex.org/W2335200603","https://openalex.org/W2183593636","https://openalex.org/W2350724007","https://openalex.org/W2080578089","https://openalex.org/W4387715848","https://openalex.org/W2179005426","https://openalex.org/W3192589309"],"abstract_inverted_index":{"Grapheme-to-Phoneme":[0],"conversion":[1],"(G2P)":[2],"is":[3,22,96,124,152],"usually":[4],"used":[5,94],"within":[6],"every":[7],"state-of-the-art":[8,90],"ASR":[9,138,160],"system":[10],"to":[11,62],"generalize":[12],"beyond":[13],"a":[14,41,109,136],"fixed":[15],"set":[16],"of":[17,35,69,111,128,143,149],"words.":[18],"Although":[19,67],"the":[20,64,70,83,93,98,115,126,141],"performance":[21,134],"typically":[23],"already":[24,73],"quite":[25],"good":[26],"(&amp;lt;":[27],"10":[28],"%":[29],"phoneme":[30],"error":[31],"rate)":[32],"and":[33,58,92],"pronunciations":[34],"important":[36],"words":[37],"are":[38,45,77,86],"checked":[39],"by":[40],"linguist,":[42],"further":[43],"improvements":[44],"still":[46],"desirable,":[47],"especially":[48],"for":[49,108,135],"end":[50],"user":[51],"customization.":[52],"In":[53],"this":[54,122],"work,":[55],"we":[56],"present":[57],"compare":[59],"five":[60],"methods/tools":[61],"tackle":[63],"G2P":[65,116,130],"task.":[66,139],"most":[68],"methods":[71],"have":[72],"been":[74],"published":[75],"and/or":[76],"available":[78],"as":[79],"open":[80],"source":[81],"software,":[82],"reported":[84],"experiments":[85],"done":[87],"on":[88,105,132],"large":[89],"tasks":[91],"software":[95],"from":[97],"actual":[99],"publications.":[100],"Besides":[101],"an":[102],"experimental":[103],"comparison":[104],"text":[106],"data":[107],"range":[110],"languages":[112],"(i.e.":[113],"measuring":[114,125],"accuracy":[117],"only),":[118],"our":[119],"focus":[120],"in":[121],"paper":[123],"effect":[127,142],"improved":[129],"modeling":[131],"LVCSR":[133],"challenging":[137],"Additionally,":[140],"using":[144],"n-Best":[145],"pronunciation":[146],"variants":[147],"instead":[148],"single":[150],"best":[151],"investigated":[153],"briefly.":[154],"Index":[155],"Terms:":[156],"grapheme-to-phoneme":[157],"conversion,":[158],"G2P,":[159],"1.":[161]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":1},{"year":2019,"cited_by_count":2},{"year":2018,"cited_by_count":4},{"year":2017,"cited_by_count":2},{"year":2016,"cited_by_count":3},{"year":2015,"cited_by_count":4},{"year":2014,"cited_by_count":3},{"year":2013,"cited_by_count":5}],"updated_date":"2026-04-05T17:49:38.594831","created_date":"2025-10-10T00:00:00"}