{"id":"https://openalex.org/W3044483536","doi":"https://doi.org/10.21437/interspeech.2020-1170","title":"Unsupervised Subword Modeling Using Autoregressive Pretraining and Cross-Lingual Phone-Aware Modeling","display_name":"Unsupervised Subword Modeling Using Autoregressive Pretraining and Cross-Lingual Phone-Aware Modeling","publication_year":2020,"publication_date":"2020-10-25","ids":{"openalex":"https://openalex.org/W3044483536","doi":"https://doi.org/10.21437/interspeech.2020-1170","mag":"3044483536"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2020-1170","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2020-1170","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":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2007.13002","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5053373302","display_name":"Siyuan Feng","orcid":"https://orcid.org/0000-0003-2531-8480"},"institutions":[{"id":"https://openalex.org/I98358874","display_name":"Delft University of Technology","ror":"https://ror.org/02e2c7k09","country_code":"NL","type":"education","lineage":["https://openalex.org/I98358874"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Siyuan Feng","raw_affiliation_strings":["Multimedia Computing Group, Delft University of Technology, Delft, the Netherlands"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Multimedia Computing Group, Delft University of Technology, Delft, the Netherlands","institution_ids":["https://openalex.org/I98358874"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5048860892","display_name":"Odette Scharenborg","orcid":"https://orcid.org/0000-0003-0693-8852"},"institutions":[{"id":"https://openalex.org/I98358874","display_name":"Delft University of Technology","ror":"https://ror.org/02e2c7k09","country_code":"NL","type":"education","lineage":["https://openalex.org/I98358874"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Odette Scharenborg","raw_affiliation_strings":["Multimedia Computing Group, Delft University of Technology, Delft, the Netherlands"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Multimedia Computing Group, Delft University of Technology, Delft, the Netherlands","institution_ids":["https://openalex.org/I98358874"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I98358874"],"apc_list":null,"apc_paid":null,"fwci":0.6505,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.75884927,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"2732","last_page":"2736"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","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/T10181","display_name":"Natural Language Processing Techniques","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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998000264167786,"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":0.9997000098228455,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8017020225524902},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.715830385684967},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6395071744918823},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.6036388874053955},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6033015847206116},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.5769801139831543},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5549967288970947},{"id":"https://openalex.org/keywords/phone","display_name":"Phone","score":0.5505268573760986},{"id":"https://openalex.org/keywords/feature-learning","display_name":"Feature learning","score":0.5264369249343872},{"id":"https://openalex.org/keywords/mandarin-chinese","display_name":"Mandarin Chinese","score":0.48108288645744324},{"id":"https://openalex.org/keywords/feature-engineering","display_name":"Feature engineering","score":0.4737092852592468},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.46499884128570557},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4495670199394226},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.43081581592559814},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.42592209577560425},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.3413575291633606}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8017020225524902},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.715830385684967},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6395071744918823},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.6036388874053955},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6033015847206116},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.5769801139831543},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5549967288970947},{"id":"https://openalex.org/C2778707766","wikidata":"https://www.wikidata.org/wiki/Q202064","display_name":"Phone","level":2,"score":0.5505268573760986},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.5264369249343872},{"id":"https://openalex.org/C138954614","wikidata":"https://www.wikidata.org/wiki/Q9192","display_name":"Mandarin Chinese","level":2,"score":0.48108288645744324},{"id":"https://openalex.org/C2778827112","wikidata":"https://www.wikidata.org/wiki/Q22245680","display_name":"Feature engineering","level":3,"score":0.4737092852592468},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.46499884128570557},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4495670199394226},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.43081581592559814},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.42592209577560425},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3413575291633606},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"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/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","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/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","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},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.21437/interspeech.2020-1170","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2020-1170","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"},{"id":"pmh:oai:arXiv.org:2007.13002","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2007.13002","pdf_url":"https://arxiv.org/pdf/2007.13002","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:tudelft.nl:uuid:525ed72a-2cc4-42bd-a753-e67c474453ae","is_oa":false,"landing_page_url":"http://resolver.tudelft.nl/uuid:525ed72a-2cc4-42bd-a753-e67c474453ae","pdf_url":null,"source":{"id":"https://openalex.org/S4306400906","display_name":"Research Repository (Delft University of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I98358874","host_organization_name":"Delft University of Technology","host_organization_lineage":["https://openalex.org/I98358874"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"conference paper"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2007.13002","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2007.13002","pdf_url":"https://arxiv.org/pdf/2007.13002","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.6700000166893005}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":34,"referenced_works":["https://openalex.org/W29952999","https://openalex.org/W162588823","https://openalex.org/W1494198834","https://openalex.org/W1522301498","https://openalex.org/W1524333225","https://openalex.org/W2012897754","https://openalex.org/W2064675550","https://openalex.org/W2085628288","https://openalex.org/W2293634267","https://openalex.org/W2399576818","https://openalex.org/W2468716020","https://openalex.org/W2514741789","https://openalex.org/W2758785877","https://openalex.org/W2786608204","https://openalex.org/W2786902352","https://openalex.org/W2787426069","https://openalex.org/W2787447541","https://openalex.org/W2842511635","https://openalex.org/W2927191280","https://openalex.org/W2949510815","https://openalex.org/W2950414763","https://openalex.org/W2963618559","https://openalex.org/W2963620343","https://openalex.org/W2963799213","https://openalex.org/W2964121744","https://openalex.org/W2972374322","https://openalex.org/W2972943112","https://openalex.org/W2995181338","https://openalex.org/W3016011332","https://openalex.org/W3016181583","https://openalex.org/W3104842308","https://openalex.org/W3125709657","https://openalex.org/W4297808394","https://openalex.org/W4300047444"],"related_works":["https://openalex.org/W3199964822","https://openalex.org/W4232132981","https://openalex.org/W4238046985","https://openalex.org/W3164948662","https://openalex.org/W4287644835","https://openalex.org/W3092281475","https://openalex.org/W3098003361","https://openalex.org/W3003242282","https://openalex.org/W3153597579","https://openalex.org/W3012824888"],"abstract_inverted_index":{"This":[0],"study":[1],"addresses":[2],"unsupervised":[3],"subword":[4],"modeling,":[5],"i.e.,<br/>learning":[6],"feature":[7,22,70,105],"representations":[8],"that":[9,100],"can":[10],"distinguish":[11],"subword<br/>units":[12],"of":[13,27,82,88,112,165],"a":[14,20,32,35,38,48,124,129,163],"language.":[15],"The":[16,91],"proposed":[17],"approach":[18],"adopts":[19],"two-stage<br/>bottleneck":[21],"(BNF)":[23],"learning":[24],"framework,":[25],"consisting":[26],"autoregressive<br/>predictive":[28],"coding":[29],"(APC)":[30],"as":[31,37,68],"front-end":[33,104],"and":[34,94],"DNNBNF<br/>model":[36],"back-end.":[39],"APC":[40,159],"pretrained":[41],"features":[42,46],"are":[43,66],"set":[44],"as<br/>input":[45],"to":[47,56,78,134,140,148,162,172],"DNN-BNF":[49],"model.":[50],"A":[51,72],"language-mismatched<br/>ASR":[52],"system":[53,145],"is":[54,77,146,156],"used":[55],"provide":[57],"cross-lingual":[58],"phone":[59],"labels":[60,119],"for<br/>DNN-BNF":[61],"model":[62],"training.":[63],"Finally,":[64],"BNFs":[65],"extracted":[67],"the<br/>subword-discriminative":[69],"representation.":[71],"second":[73],"aim":[74],"of<br/>this":[75],"work":[76],"investigate":[79],"the":[80,95,110,113,135,153],"robustness":[81],"our":[83],"approach\u2019s":[84],"effectiveness<br/>to":[85],"different":[86],"amounts":[87],"training":[89,149,154],"data.":[90],"results":[92],"on<br/>Libri-light":[93],"ZeroSpeech":[96],"2017":[97],"databases":[98],"show":[99],"APC<br/>is":[101],"effective":[102],"in":[103],"pretraining.":[106],"Our":[107,144],"whole":[108],"system<br/>outperforms":[109],"state":[111],"art":[114],"on":[115],"both":[116],"databases.":[117],"Cross-lingual<br/>phone":[118],"for":[120],"English":[121],"data":[122,150,155],"by":[123],"Dutch":[125],"ASR":[126],"outperform":[127],"those<br/>by":[128],"Mandarin":[130,141],"ASR,":[131],"possibly":[132],"linked":[133],"larger":[136],"similarity":[137],"of<br/>Dutch":[138],"compared":[139],"with":[142],"English.":[143],"less<br/>sensitive":[147],"amount":[151],"when":[152],"over<br/>50":[157],"hours.":[158],"pretraining":[160],"leads":[161],"reduction":[164],"needed":[166],"training<br/>material":[167],"from":[168],"over":[169],"5,000":[170],"hours":[171,175],"around":[173],"200":[174],"with<br/>little":[176],"performance":[177],"degradation.":[178]},"counts_by_year":[{"year":2021,"cited_by_count":4},{"year":2020,"cited_by_count":1}],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2020-07-29T00:00:00"}