{"id":"https://openalex.org/W3011582891","doi":"https://doi.org/10.1109/apsipaasc47483.2019.9023010","title":"Unsupervised Pronunciation Fluency Scoring by infoGan","display_name":"Unsupervised Pronunciation Fluency Scoring by infoGan","publication_year":2019,"publication_date":"2019-11-01","ids":{"openalex":"https://openalex.org/W3011582891","doi":"https://doi.org/10.1109/apsipaasc47483.2019.9023010","mag":"3011582891"},"language":"en","primary_location":{"id":"doi:10.1109/apsipaasc47483.2019.9023010","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc47483.2019.9023010","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","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/A5100549754","display_name":"Wenwei Dong","orcid":null},"institutions":[{"id":"https://openalex.org/I115212828","display_name":"Beijing Language and Culture University","ror":"https://ror.org/03te2zs36","country_code":"CN","type":"education","lineage":["https://openalex.org/I115212828"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Wenwei Dong","raw_affiliation_strings":["Beijing Advanced Innovation Center for Language Resources, Beijing Language and Culture University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beijing Advanced Innovation Center for Language Resources, Beijing Language and Culture University, Beijing, China","institution_ids":["https://openalex.org/I115212828"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006684220","display_name":"Yanlu Xie","orcid":"https://orcid.org/0000-0001-6765-4808"},"institutions":[{"id":"https://openalex.org/I115212828","display_name":"Beijing Language and Culture University","ror":"https://ror.org/03te2zs36","country_code":"CN","type":"education","lineage":["https://openalex.org/I115212828"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanlu Xie","raw_affiliation_strings":["Beijing Advanced Innovation Center for Language Resources, Beijing Language and Culture University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beijing Advanced Innovation Center for Language Resources, Beijing Language and Culture University, Beijing, China","institution_ids":["https://openalex.org/I115212828"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101084840","display_name":"Binghuai Lin","orcid":null},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Binghuai Lin","raw_affiliation_strings":["MIG Tencent Science and Technology Ltd., Beijing, China"],"affiliations":[{"raw_affiliation_string":"MIG Tencent Science and Technology Ltd., Beijing, China","institution_ids":["https://openalex.org/I2250653659"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5100549754"],"corresponding_institution_ids":["https://openalex.org/I115212828"],"apc_list":null,"apc_paid":null,"fwci":0.2886,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.69167574,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":"1","issue":null,"first_page":"1020","last_page":"1023"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9984999895095825,"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":0.9984999895095825,"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":0.9884999990463257,"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/T12031","display_name":"Speech and dialogue systems","score":0.9753999710083008,"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":0.8392099738121033},{"id":"https://openalex.org/keywords/fluency","display_name":"Fluency","score":0.7343331575393677},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7047214508056641},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5161095857620239},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5118938684463501},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4059472680091858},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.24341335892677307}],"concepts":[{"id":"https://openalex.org/C2780844864","wikidata":"https://www.wikidata.org/wiki/Q184377","display_name":"Pronunciation","level":2,"score":0.8392099738121033},{"id":"https://openalex.org/C2777413886","wikidata":"https://www.wikidata.org/wiki/Q3276013","display_name":"Fluency","level":2,"score":0.7343331575393677},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7047214508056641},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5161095857620239},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5118938684463501},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4059472680091858},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.24341335892677307},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/apsipaasc47483.2019.9023010","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc47483.2019.9023010","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","raw_type":"proceedings-article"},{"id":"mag:3041083897","is_oa":false,"landing_page_url":"https://jglobal.jst.go.jp/en/detail?JGLOBAL_ID=202002250478357796","pdf_url":null,"source":{"id":"https://openalex.org/S4306512817","display_name":"IEEE Conference Proceedings","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":"IEEE Conference Proceedings","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.8399999737739563,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W1496430420","https://openalex.org/W1567526008","https://openalex.org/W2027993708","https://openalex.org/W2091856355","https://openalex.org/W2295721380","https://openalex.org/W2401896499","https://openalex.org/W2768986418","https://openalex.org/W2805034849","https://openalex.org/W2888977743","https://openalex.org/W2889415062","https://openalex.org/W2889562042","https://openalex.org/W2963226019","https://openalex.org/W6718140377"],"related_works":["https://openalex.org/W2183593636","https://openalex.org/W2350724007","https://openalex.org/W2355751417","https://openalex.org/W2423284978","https://openalex.org/W2083922162","https://openalex.org/W2000075989","https://openalex.org/W4220683390","https://openalex.org/W2365169615","https://openalex.org/W2776838583","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Pronunciation":[0],"fluency":[1,95],"scoring":[2,32],"(PFS)":[3],"is":[4,38,62],"a":[5,77],"primary":[6],"task":[7],"in":[8],"computer-aided":[9],"second":[10],"language":[11],"(L2)":[12],"learning.":[13],"Most":[14],"of":[15],"existing":[16],"PFS":[17],"algorithms":[18],"are":[19,27,73],"based":[20],"on":[21],"supervised":[22],"learning,":[23],"where":[24,58],"human-labeled":[25],"scores":[26],"used":[28,74],"to":[29,43,48,64,75],"train":[30],"the":[31,35],"model.":[33],"However,":[34],"human":[36],"labeling":[37],"rather":[39],"costly":[40],"and":[41,69,82],"tends":[42],"be":[44],"biased.":[45],"In":[46],"order":[47],"tackle":[49],"this":[50,88],"problem,":[51],"we":[52],"propose":[53],"an":[54,59],"unsupervised":[55],"learning":[56],"approach,":[57],"infoGan":[60],"model":[61],"constructed":[63],"infer":[65],"latent":[66],"speech":[67],"codes,":[68],"then":[70],"these":[71],"codes":[72],"build":[76],"classifier":[78,90],"that":[79,87],"distinguishes":[80],"native":[81],"foreign":[83],"speech.":[84],"We":[85],"found":[86],"native-foreign":[89],"can":[91],"generate":[92],"good":[93],"utterance-based":[94],"scores.":[96]},"counts_by_year":[{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":1}],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2025-10-10T00:00:00"}