{"id":"https://openalex.org/W3096674206","doi":"https://doi.org/10.21437/interspeech.2020-2953","title":"Context-Aware Goodness of Pronunciation for Computer-Assisted Pronunciation Training","display_name":"Context-Aware Goodness of Pronunciation for Computer-Assisted Pronunciation Training","publication_year":2020,"publication_date":"2020-10-25","ids":{"openalex":"https://openalex.org/W3096674206","doi":"https://doi.org/10.21437/interspeech.2020-2953","mag":"3096674206"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2020-2953","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2020-2953","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2020","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/A5008336983","display_name":"Jiatong Shi","orcid":"https://orcid.org/0000-0002-9050-8304"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Jiatong Shi","raw_affiliation_strings":["Department of Computer Science, Johns Hopkins University, U.S.A"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, Johns Hopkins University, U.S.A","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071652041","display_name":"Nan Huo","orcid":"https://orcid.org/0000-0002-9305-3309"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nan Huo","raw_affiliation_strings":["Department of Computer Science, Johns Hopkins University, U.S.A"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, Johns Hopkins University, U.S.A","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5009985839","display_name":"Qin Jin","orcid":"https://orcid.org/0000-0001-6486-6020"},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qin Jin","raw_affiliation_strings":["School of Information, Renmin University of China, P.R.China"],"affiliations":[{"raw_affiliation_string":"School of Information, Renmin University of China, P.R.China","institution_ids":["https://openalex.org/I78988378"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5008336983"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":4.9396,"has_fulltext":false,"cited_by_count":50,"citation_normalized_percentile":{"value":0.96008498,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"3057","last_page":"3061"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9968000054359436,"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":0.9968000054359436,"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/T10403","display_name":"Phonetics and Phonology Research","score":0.9879999756813049,"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/T12031","display_name":"Speech and dialogue systems","score":0.9377999901771545,"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/pronunciation","display_name":"Pronunciation","score":0.9486095905303955},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7307944893836975},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.643470823764801},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5123652219772339},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5026390552520752},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.45874011516571045},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.2455829381942749},{"id":"https://openalex.org/keywords/geology","display_name":"Geology","score":0.04497748613357544}],"concepts":[{"id":"https://openalex.org/C2780844864","wikidata":"https://www.wikidata.org/wiki/Q184377","display_name":"Pronunciation","level":2,"score":0.9486095905303955},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7307944893836975},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.643470823764801},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5123652219772339},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5026390552520752},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.45874011516571045},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.2455829381942749},{"id":"https://openalex.org/C127313418","wikidata":"https://www.wikidata.org/wiki/Q1069","display_name":"Geology","level":0,"score":0.04497748613357544},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2020-2953","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2020-2953","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2020","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.699999988079071}],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":31,"referenced_works":["https://openalex.org/W32018921","https://openalex.org/W334543181","https://openalex.org/W615517874","https://openalex.org/W1494198834","https://openalex.org/W1503327452","https://openalex.org/W1524333225","https://openalex.org/W1533861849","https://openalex.org/W1985084205","https://openalex.org/W2026408911","https://openalex.org/W2109191021","https://openalex.org/W2119607964","https://openalex.org/W2147768505","https://openalex.org/W2295209184","https://openalex.org/W2398423749","https://openalex.org/W2398741870","https://openalex.org/W2402146185","https://openalex.org/W2402640444","https://openalex.org/W2405144467","https://openalex.org/W2408752745","https://openalex.org/W2514741789","https://openalex.org/W2515321869","https://openalex.org/W2524467597","https://openalex.org/W2552635739","https://openalex.org/W2748565458","https://openalex.org/W2749298174","https://openalex.org/W2888030130","https://openalex.org/W2888867175","https://openalex.org/W2964089206","https://openalex.org/W2972940133","https://openalex.org/W2978725006","https://openalex.org/W4385245566"],"related_works":["https://openalex.org/W2375756441","https://openalex.org/W1990362134","https://openalex.org/W2789919619","https://openalex.org/W4223932119","https://openalex.org/W2293457016","https://openalex.org/W2384728334","https://openalex.org/W2547369883","https://openalex.org/W3169305685","https://openalex.org/W2351428524","https://openalex.org/W2368779261"],"abstract_inverted_index":{"Mispronunciation":[0],"detection":[1,14,177],"is":[2,144],"an":[3],"essential":[4],"component":[5],"of":[6,27,103],"the":[7,52,65,70,83,100,111,115,126,135,148,152,168,172],"Computer-Assisted":[8],"Pronunciation":[9,28,104],"Training":[10],"(CAPT)":[11],"systems.State-of-the-art":[12],"mispronunciation":[13,176],"models":[15,38],"use":[16,59],"Deep":[17],"Neural":[18],"Networks":[19],"(DNN)":[20],"for":[21,32,61],"acoustic":[22],"modeling,":[23],"and":[24,57,114,130,163,174],"a":[25,138],"Goodness":[26,102],"(GOP)":[29],"based":[30,36,140],"algorithm":[31],"pronunciation":[33],"scoring.However,":[34],"GOP":[35,169],"scoring":[37,106,153,156],"have":[39],"two":[40,108],"major":[41],"limitations:":[42],"i.e.,":[43],"(i)":[44],"They":[45,73],"depend":[46],"on":[47,76,171],"forced":[48],"alignment":[49],"which":[50,63,79],"splits":[51],"speech":[53],"into":[54,120,151],"phonetic":[55,77,141],"segments":[56],"independently":[58],"them":[60,132],"scoring,":[62],"neglects":[64],"transitions":[66,127],"between":[67,128],"phonemes":[68,87,129],"within":[69],"segment;":[71],"(ii)":[72],"only":[74],"focus":[75],"segments,":[78],"fails":[80],"to":[81,133,146],"consider":[82],"context":[84],"effects":[85],"across":[86],"(such":[88],"as":[89],"liaison,":[90],"omission,":[91],"incomplete":[92],"plosive":[93],"sound,":[94],"etc.).In":[95],"this":[96],"work,":[97],"we":[98],"propose":[99],"Context-aware":[101],"(CaGOP)":[105],"model.Particularly,":[107],"factors":[109],"namely":[110],"transition":[112,123],"factor":[113,117,124,150],"duration":[116,142,149],"are":[118],"injected":[119],"CaGOP":[121],"scoring.The":[122],"identifies":[125],"applies":[131],"weight":[134],"frame-wise":[136],"GOP.Moreover,":[137],"self-attention":[139],"modeling":[143],"proposed":[145,155],"introduce":[147],"model.The":[154],"model":[157,170],"significantly":[158],"outperforms":[159],"baselines,":[160],"achieving":[161],"20%":[162],"12%":[164],"relative":[165],"improvement":[166],"over":[167],"phoneme-level":[173],"sentence-level":[175],"respectively.":[178]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":7},{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":18},{"year":2022,"cited_by_count":10},{"year":2021,"cited_by_count":8}],"updated_date":"2026-03-15T09:29:46.208133","created_date":"2025-10-10T00:00:00"}