{"id":"https://openalex.org/W2936078256","doi":"https://doi.org/10.1109/icassp.2019.8682586","title":"The Speechtransformer for Large-scale Mandarin Chinese Speech Recognition","display_name":"The Speechtransformer for Large-scale Mandarin Chinese Speech Recognition","publication_year":2019,"publication_date":"2019-04-17","ids":{"openalex":"https://openalex.org/W2936078256","doi":"https://doi.org/10.1109/icassp.2019.8682586","mag":"2936078256"},"language":"en","primary_location":{"id":"doi:10.1109/icassp.2019.8682586","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2019.8682586","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5101995701","display_name":"Yuanyuan Zhao","orcid":"https://orcid.org/0000-0002-6107-2185"},"institutions":[{"id":"https://openalex.org/I2801745840","display_name":"Kwai Chung Hospital","ror":"https://ror.org/05kz7bw59","country_code":"CN","type":"healthcare","lineage":["https://openalex.org/I1294586568","https://openalex.org/I2801745840"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuanyuan zhao","raw_affiliation_strings":["Kwai, Beijing, P.R. China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kwai, Beijing, P.R. China","institution_ids":["https://openalex.org/I2801745840"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028530320","display_name":"Jie Li","orcid":"https://orcid.org/0000-0002-7075-4145"},"institutions":[{"id":"https://openalex.org/I2801745840","display_name":"Kwai Chung Hospital","ror":"https://ror.org/05kz7bw59","country_code":"CN","type":"healthcare","lineage":["https://openalex.org/I1294586568","https://openalex.org/I2801745840"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jie Li","raw_affiliation_strings":["Kwai, Beijing, P.R. China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kwai, Beijing, P.R. China","institution_ids":["https://openalex.org/I2801745840"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100658019","display_name":"Xiaorui Wang","orcid":"https://orcid.org/0000-0001-9633-1418"},"institutions":[{"id":"https://openalex.org/I2801745840","display_name":"Kwai Chung Hospital","ror":"https://ror.org/05kz7bw59","country_code":"CN","type":"healthcare","lineage":["https://openalex.org/I1294586568","https://openalex.org/I2801745840"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaorui Wang","raw_affiliation_strings":["Kwai, Beijing, P.R. China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kwai, Beijing, P.R. China","institution_ids":["https://openalex.org/I2801745840"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101886099","display_name":"Yan Li","orcid":"https://orcid.org/0000-0003-3566-0992"},"institutions":[{"id":"https://openalex.org/I2801745840","display_name":"Kwai Chung Hospital","ror":"https://ror.org/05kz7bw59","country_code":"CN","type":"healthcare","lineage":["https://openalex.org/I1294586568","https://openalex.org/I2801745840"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yan Li","raw_affiliation_strings":["Kwai, Beijing, P.R. China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kwai, Beijing, P.R. China","institution_ids":["https://openalex.org/I2801745840"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I2801745840"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":90,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7095","last_page":"7099"},"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9979000091552734,"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":0.996999979019165,"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/computer-science","display_name":"Computer science","score":0.8108588457107544},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.7721245288848877},{"id":"https://openalex.org/keywords/mandarin-chinese","display_name":"Mandarin Chinese","score":0.7540147304534912},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.5971741080284119},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5955233573913574},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5934594869613647},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.5122212767601013},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.47037404775619507},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.46295031905174255},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4332996606826782},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4172341525554657},{"id":"https://openalex.org/keywords/speech-enhancement","display_name":"Speech enhancement","score":0.41235750913619995},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.36467790603637695},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.12777680158615112},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.06772184371948242}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8108588457107544},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.7721245288848877},{"id":"https://openalex.org/C138954614","wikidata":"https://www.wikidata.org/wiki/Q9192","display_name":"Mandarin Chinese","level":2,"score":0.7540147304534912},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.5971741080284119},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5955233573913574},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5934594869613647},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5122212767601013},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.47037404775619507},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.46295031905174255},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4332996606826782},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4172341525554657},{"id":"https://openalex.org/C2776182073","wikidata":"https://www.wikidata.org/wiki/Q7575395","display_name":"Speech enhancement","level":3,"score":0.41235750913619995},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.36467790603637695},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.12777680158615112},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.06772184371948242},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"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":1,"locations":[{"id":"doi:10.1109/icassp.2019.8682586","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2019.8682586","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":39,"referenced_works":["https://openalex.org/W648786980","https://openalex.org/W854541894","https://openalex.org/W1586532344","https://openalex.org/W1779452081","https://openalex.org/W1828163288","https://openalex.org/W1922655562","https://openalex.org/W2183341477","https://openalex.org/W2327501763","https://openalex.org/W2514741789","https://openalex.org/W2515439472","https://openalex.org/W2525778437","https://openalex.org/W2545177271","https://openalex.org/W2729190387","https://openalex.org/W2748163763","https://openalex.org/W2802023636","https://openalex.org/W2884561390","https://openalex.org/W2889163603","https://openalex.org/W2892009249","https://openalex.org/W2962742956","https://openalex.org/W2962778134","https://openalex.org/W2962824709","https://openalex.org/W2962826786","https://openalex.org/W2963070863","https://openalex.org/W2963242190","https://openalex.org/W2963403868","https://openalex.org/W2963483399","https://openalex.org/W2963850025","https://openalex.org/W2963920996","https://openalex.org/W2964089206","https://openalex.org/W4297797495","https://openalex.org/W4385245566","https://openalex.org/W6621543089","https://openalex.org/W6623517193","https://openalex.org/W6635078382","https://openalex.org/W6638005537","https://openalex.org/W6638749077","https://openalex.org/W6640090968","https://openalex.org/W6728910023","https://openalex.org/W6739901393"],"related_works":["https://openalex.org/W2163874654","https://openalex.org/W1566315437","https://openalex.org/W2401089611","https://openalex.org/W2594897229","https://openalex.org/W4221142855","https://openalex.org/W2151348424","https://openalex.org/W2050138804","https://openalex.org/W4290708361","https://openalex.org/W2129812225","https://openalex.org/W2523799048"],"abstract_inverted_index":{"Attention-based":[0],"sequence-to-sequence":[1],"architectures":[2],"have":[3],"made":[4],"great":[5],"progress":[6],"in":[7,28,120],"the":[8,52,57,78,83,103,113,147],"speech":[9,24,41],"recognition":[10,25,42],"task.":[11],"The":[12,86],"SpeechTransformer,":[13],"a":[14,37,65,109,129,143],"no-recurrence":[15],"encoder-decoder":[16],"architecture,":[17],"has":[18],"shown":[19,72],"promising":[20],"results":[21],"on":[22,36,122],"small-scale":[23],"data":[26],"sets":[27],"previous":[29],"works.":[30],"In":[31],"this":[32],"paper,":[33],"we":[34],"focus":[35],"large-scale":[38],"Mandarin":[39],"Chinese":[40],"task":[43],"and":[44,54,93,140],"propose":[45],"three":[46],"optimization":[47],"strategies":[48,89],"to":[49,63,75,101,128],"further":[50],"improve":[51],"performance":[53],"efficiency":[55,80],"of":[56],"SpeechTransformer.":[58],"Our":[59],"first":[60],"improvement":[61],"is":[62,71,135],"use":[64],"much":[66],"lower":[67],"frame":[68],"rate,":[69],"which":[70,96,134],"very":[73,99],"beneficial":[74],"not":[76],"only":[77],"computation":[79],"but":[81],"also":[82],"model":[84],"performance.":[85],"other":[87],"two":[88],"are":[90,97],"scheduled":[91],"sampling":[92],"focal":[94],"loss,":[95],"both":[98],"effective":[100],"reduce":[102],"character":[104],"error":[105],"rate":[106],"(CER).":[107],"On":[108],"8,000":[110],"hours":[111],"task,":[112],"proposed":[114],"improvements":[115],"yield":[116],"10.8%-26.1%":[117],"relative":[118,153],"gain":[119],"CER":[121,154],"four":[123],"different":[124],"test":[125],"sets.":[126],"Compared":[127],"strong":[130],"hybrid":[131],"TDNN-LSTM":[132],"system,":[133],"trained":[136],"with":[137,142],"LF-MMI":[138],"criterion":[139],"decoded":[141],"large":[144],"4-gram":[145],"LM,":[146],"final":[148],"optimized":[149],"Speech-Transformer":[150],"gives":[151],"12.2%-19.1%":[152],"reduction":[155],"without":[156],"any":[157],"explicit":[158],"language":[159],"models.":[160]},"counts_by_year":[{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":11},{"year":2023,"cited_by_count":6},{"year":2022,"cited_by_count":10},{"year":2021,"cited_by_count":24},{"year":2020,"cited_by_count":26},{"year":2019,"cited_by_count":8}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}