{"id":"https://openalex.org/W4210423654","doi":"https://doi.org/10.1109/asru51503.2021.9688053","title":"Decoupling Recognition and Transcription in Mandarin ASR","display_name":"Decoupling Recognition and Transcription in Mandarin ASR","publication_year":2021,"publication_date":"2021-12-13","ids":{"openalex":"https://openalex.org/W4210423654","doi":"https://doi.org/10.1109/asru51503.2021.9688053"},"language":"en","primary_location":{"id":"doi:10.1109/asru51503.2021.9688053","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru51503.2021.9688053","pdf_url":null,"source":{"id":"https://openalex.org/S4363606113","display_name":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","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":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","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/A5043406907","display_name":"Jiahong Yuan","orcid":"https://orcid.org/0009-0008-2162-7167"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Jiahong Yuan","raw_affiliation_strings":["Baidu Research USA"],"affiliations":[{"raw_affiliation_string":"Baidu Research USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013075695","display_name":"Xingyu Cai","orcid":"https://orcid.org/0000-0003-1537-7161"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xingyu Cai","raw_affiliation_strings":["Baidu Research USA"],"affiliations":[{"raw_affiliation_string":"Baidu Research USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072814956","display_name":"Dongji Gao","orcid":"https://orcid.org/0009-0006-8885-3084"},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dongji Gao","raw_affiliation_strings":["Johns Hopkins University"],"affiliations":[{"raw_affiliation_string":"Johns Hopkins University","institution_ids":["https://openalex.org/I145311948"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034431550","display_name":"Renjie Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Renjie Zheng","raw_affiliation_strings":["Baidu Research USA"],"affiliations":[{"raw_affiliation_string":"Baidu Research USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101440599","display_name":"Liang Huang","orcid":"https://orcid.org/0009-0007-5846-849X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang Huang","raw_affiliation_strings":["Baidu Research USA"],"affiliations":[{"raw_affiliation_string":"Baidu Research USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5016543371","display_name":"Kenneth Church","orcid":"https://orcid.org/0000-0001-8378-6069"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kenneth Church","raw_affiliation_strings":["Baidu Research USA"],"affiliations":[{"raw_affiliation_string":"Baidu Research USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5043406907"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.0053,"has_fulltext":false,"cited_by_count":12,"citation_normalized_percentile":{"value":0.80370163,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1019","last_page":"1025"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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":0.9998999834060669,"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":0.9979000091552734,"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/T11309","display_name":"Music and Audio Processing","score":0.996999979019165,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/pinyin","display_name":"Pinyin","score":0.9758220911026001},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7722922563552856},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.7577154040336609},{"id":"https://openalex.org/keywords/transcription","display_name":"Transcription (linguistics)","score":0.6984754800796509},{"id":"https://openalex.org/keywords/mandarin-chinese","display_name":"Mandarin Chinese","score":0.6501030921936035},{"id":"https://openalex.org/keywords/factoring","display_name":"Factoring","score":0.5959112644195557},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.487822562456131},{"id":"https://openalex.org/keywords/chinese-speech-synthesis","display_name":"Chinese speech synthesis","score":0.4780690670013428},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.4600536525249481},{"id":"https://openalex.org/keywords/chinese-characters","display_name":"Chinese characters","score":0.4483892321586609},{"id":"https://openalex.org/keywords/decoupling","display_name":"Decoupling (probability)","score":0.42526674270629883},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3566295802593231},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.3110353946685791},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.27845507860183716},{"id":"https://openalex.org/keywords/speech-technology","display_name":"Speech technology","score":0.16540685296058655},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08211040496826172}],"concepts":[{"id":"https://openalex.org/C2781095461","wikidata":"https://www.wikidata.org/wiki/Q42222","display_name":"Pinyin","level":3,"score":0.9758220911026001},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7722922563552856},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.7577154040336609},{"id":"https://openalex.org/C179926584","wikidata":"https://www.wikidata.org/wiki/Q207714","display_name":"Transcription (linguistics)","level":2,"score":0.6984754800796509},{"id":"https://openalex.org/C138954614","wikidata":"https://www.wikidata.org/wiki/Q9192","display_name":"Mandarin Chinese","level":2,"score":0.6501030921936035},{"id":"https://openalex.org/C177225278","wikidata":"https://www.wikidata.org/wiki/Q192674","display_name":"Factoring","level":2,"score":0.5959112644195557},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.487822562456131},{"id":"https://openalex.org/C73411735","wikidata":"https://www.wikidata.org/wiki/Q16369","display_name":"Chinese speech synthesis","level":4,"score":0.4780690670013428},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.4600536525249481},{"id":"https://openalex.org/C2781051154","wikidata":"https://www.wikidata.org/wiki/Q8201","display_name":"Chinese characters","level":2,"score":0.4483892321586609},{"id":"https://openalex.org/C205606062","wikidata":"https://www.wikidata.org/wiki/Q5249645","display_name":"Decoupling (probability)","level":2,"score":0.42526674270629883},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3566295802593231},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3110353946685791},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.27845507860183716},{"id":"https://openalex.org/C504749915","wikidata":"https://www.wikidata.org/wiki/Q9010971","display_name":"Speech technology","level":3,"score":0.16540685296058655},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08211040496826172},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C10138342","wikidata":"https://www.wikidata.org/wiki/Q43015","display_name":"Finance","level":1,"score":0.0},{"id":"https://openalex.org/C133731056","wikidata":"https://www.wikidata.org/wiki/Q4917288","display_name":"Control engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asru51503.2021.9688053","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru51503.2021.9688053","pdf_url":null,"source":{"id":"https://openalex.org/S4363606113","display_name":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","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":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.7400000095367432,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":81,"referenced_works":["https://openalex.org/W50577135","https://openalex.org/W108550785","https://openalex.org/W1571745681","https://openalex.org/W1571901760","https://openalex.org/W2127141656","https://openalex.org/W2167513360","https://openalex.org/W2514969556","https://openalex.org/W2755682845","https://openalex.org/W2900819823","https://openalex.org/W2963242190","https://openalex.org/W2963850025","https://openalex.org/W2964002616","https://openalex.org/W2972324550","https://openalex.org/W2972686346","https://openalex.org/W2990906560","https://openalex.org/W3015208154","https://openalex.org/W3021459083","https://openalex.org/W3023126978","https://openalex.org/W3023751411","https://openalex.org/W3025165719","https://openalex.org/W3036601975","https://openalex.org/W3095783102","https://openalex.org/W3097345069","https://openalex.org/W3097438275","https://openalex.org/W3097625183","https://openalex.org/W3097777922","https://openalex.org/W3097829404","https://openalex.org/W3097874139","https://openalex.org/W3109196171","https://openalex.org/W3111562797","https://openalex.org/W3113594615","https://openalex.org/W3140941716","https://openalex.org/W3141854550","https://openalex.org/W3144345593","https://openalex.org/W3145511196","https://openalex.org/W3148986242","https://openalex.org/W3150269179","https://openalex.org/W3151526698","https://openalex.org/W3152519008","https://openalex.org/W3154511834","https://openalex.org/W3156902660","https://openalex.org/W3160345865","https://openalex.org/W3160523328","https://openalex.org/W3160622492","https://openalex.org/W3162016613","https://openalex.org/W3162249256","https://openalex.org/W3163487528","https://openalex.org/W3180992661","https://openalex.org/W3192424912","https://openalex.org/W3197140813","https://openalex.org/W3197478142","https://openalex.org/W3197522112","https://openalex.org/W3197737164","https://openalex.org/W3198528147","https://openalex.org/W4225644313","https://openalex.org/W4246002569","https://openalex.org/W4385245566","https://openalex.org/W6602079780","https://openalex.org/W6604322312","https://openalex.org/W6776088777","https://openalex.org/W6777158299","https://openalex.org/W6777478455","https://openalex.org/W6777859476","https://openalex.org/W6780218876","https://openalex.org/W6784364040","https://openalex.org/W6784599769","https://openalex.org/W6785049911","https://openalex.org/W6785107891","https://openalex.org/W6786266511","https://openalex.org/W6786271521","https://openalex.org/W6787040858","https://openalex.org/W6787141514","https://openalex.org/W6791959309","https://openalex.org/W6792376965","https://openalex.org/W6793678319","https://openalex.org/W6793772276","https://openalex.org/W6793959113","https://openalex.org/W6794876948","https://openalex.org/W6794920533","https://openalex.org/W6800486944","https://openalex.org/W6801377430"],"related_works":["https://openalex.org/W4385493412","https://openalex.org/W3159297355","https://openalex.org/W1575804004","https://openalex.org/W2386864954","https://openalex.org/W380382270","https://openalex.org/W2547767923","https://openalex.org/W1788792945","https://openalex.org/W2529685800","https://openalex.org/W1932828703","https://openalex.org/W2372811119"],"abstract_inverted_index":{"Much":[0],"of":[1,56,59],"the":[2,18,63,78,81],"recent":[3],"literature":[4],"on":[5,77,85],"automatic":[6],"speech":[7],"recognition":[8],"(ASR)":[9],"is":[10,21,53],"taking":[11],"an":[12],"end-to-end":[13],"approach.":[14],"Unlike":[15],"English":[16],"where":[17,51],"writing":[19],"system":[20,55],"closely":[22],"related":[23],"to":[24],"sound,":[25],"Chinese":[26],"characters":[27],"(Hanzi)":[28],"represent":[29],"meaning,":[30],"not":[31],"sound.":[32],"We":[33],"propose":[34],"factoring":[35],"audio":[36,43,64],"\u2192":[37,44,49,65],"Hanzi":[38,66],"into":[39],"two":[40],"sub-tasks:":[41],"(1)":[42],"Pinyin":[45,48,52],"and":[46],"(2)":[47],"Hanzi,":[50],"a":[54],"phonetic":[57],"transcription":[58],"standard":[60],"Chinese.":[61],"Factoring":[62],"task":[67],"in":[68],"this":[69,86],"way":[70],"achieves":[71],"3.9%":[72],"CER":[73],"(character":[74],"error":[75],"rate)":[76],"Aishell-1":[79],"corpus,":[80],"best":[82],"result":[83],"reported":[84],"dataset":[87],"so":[88],"far.":[89]},"counts_by_year":[{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}