{"id":"https://openalex.org/W3161571818","doi":"https://doi.org/10.1109/icassp39728.2021.9415004","title":"Convolutional Dropout and Wordpiece Augmentation for End-to-End Speech Recognition","display_name":"Convolutional Dropout and Wordpiece Augmentation for End-to-End Speech Recognition","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3161571818","doi":"https://doi.org/10.1109/icassp39728.2021.9415004","mag":"3161571818"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9415004","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9415004","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 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/A5034399968","display_name":"Hainan Xu","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hainan Xu","raw_affiliation_strings":["Google Inc,New York,NY,U.S.A","Google Inc, New York, NY, U.S.A"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Inc,New York,NY,U.S.A","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc, New York, NY, U.S.A","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101673889","display_name":"Yinghui Huang","orcid":"https://orcid.org/0000-0003-0607-1507"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yinghui Huang","raw_affiliation_strings":["Google Inc,New York,NY,U.S.A","Google Inc, New York, NY, U.S.A"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Inc,New York,NY,U.S.A","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc, New York, NY, U.S.A","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061165176","display_name":"Yun Zhu","orcid":"https://orcid.org/0000-0001-6734-8202"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yun Zhu","raw_affiliation_strings":["Google Inc,New York,NY,U.S.A","Google Inc, New York, NY, U.S.A"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Inc,New York,NY,U.S.A","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc, New York, NY, U.S.A","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015927589","display_name":"Kartik Audhkhasi","orcid":"https://orcid.org/0000-0002-2340-1144"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kartik Audhkhasi","raw_affiliation_strings":["Google Inc,New York,NY,U.S.A","Google Inc, New York, NY, U.S.A"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Inc,New York,NY,U.S.A","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc, New York, NY, U.S.A","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5071715737","display_name":"Bhuvana Ramabhadran","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bhuvana Ramabhadran","raw_affiliation_strings":["Google Inc,New York,NY,U.S.A","Google Inc, New York, NY, U.S.A"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Inc,New York,NY,U.S.A","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc, New York, NY, U.S.A","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":"15","issue":null,"first_page":"5984","last_page":"5988"},"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":0.9991000294685364,"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/T11309","display_name":"Music and Audio Processing","score":0.9965000152587891,"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/end-to-end-principle","display_name":"End-to-end principle","score":0.8341842293739319},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7150796055793762},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6706375479698181},{"id":"https://openalex.org/keywords/dropout","display_name":"Dropout (neural networks)","score":0.6424431800842285},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.35108572244644165},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.07677879929542542}],"concepts":[{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.8341842293739319},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7150796055793762},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6706375479698181},{"id":"https://openalex.org/C2776145597","wikidata":"https://www.wikidata.org/wiki/Q25339462","display_name":"Dropout (neural networks)","level":2,"score":0.6424431800842285},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35108572244644165},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.07677879929542542}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9415004","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9415004","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.5899999737739563}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W4919037","https://openalex.org/W1494198834","https://openalex.org/W1828163288","https://openalex.org/W2095705004","https://openalex.org/W2407080277","https://openalex.org/W2525778437","https://openalex.org/W2606722458","https://openalex.org/W2617258110","https://openalex.org/W2890166761","https://openalex.org/W2928941594","https://openalex.org/W2936774411","https://openalex.org/W2938348542","https://openalex.org/W2941814890","https://openalex.org/W2952634764","https://openalex.org/W2962784628","https://openalex.org/W2963250244","https://openalex.org/W2963403868","https://openalex.org/W2975381464","https://openalex.org/W2996159613","https://openalex.org/W3015284407","https://openalex.org/W3016010032","https://openalex.org/W3025165719","https://openalex.org/W3026041220","https://openalex.org/W3035207248","https://openalex.org/W3036601975","https://openalex.org/W3097777922","https://openalex.org/W3099782249","https://openalex.org/W4385245566","https://openalex.org/W6600213771","https://openalex.org/W6638575559","https://openalex.org/W6638749077","https://openalex.org/W6674330103","https://openalex.org/W6713762819","https://openalex.org/W6739901393","https://openalex.org/W6754484989","https://openalex.org/W6760633627","https://openalex.org/W6761563299","https://openalex.org/W6761715637","https://openalex.org/W6768080748","https://openalex.org/W6769361421","https://openalex.org/W6780218876"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W3082178636","https://openalex.org/W2151749779","https://openalex.org/W2782041652","https://openalex.org/W3179968364","https://openalex.org/W2612657834","https://openalex.org/W2392157706","https://openalex.org/W2599192953"],"abstract_inverted_index":{"Regularization":[0],"and":[1,118],"data":[2,93],"augmentation":[3,94],"are":[4],"crucial":[5],"to":[6,80],"training":[7,112],"end-to-end":[8],"automatic":[9],"speech":[10,115],"recognition":[11,116],"systems.":[12],"Dropout":[13],"is":[14,46],"a":[15,29,35,49,92],"popular":[16],"regularization":[17],"technique,":[18],"which":[19],"operates":[20],"on":[21,123],"each":[22,43],"neuron":[23,54],"independently":[24],"by":[25,99],"multiplying":[26],"it":[27],"with":[28,48,70],"Bernoulli":[30],"random":[31,88],"variable.":[32],"We":[33,59,105],"propose":[34],"generalization":[36],"of":[37,53,68,74,113],"dropout,":[38,82],"called":[39],"\"convolutional":[40],"dropout\",":[41],"where":[42],"neuron\u2019s":[44],"activation":[45],"replaced":[47],"randomly-weighted":[50],"linear":[51],"combination":[52],"values":[55],"in":[56,101],"its":[57],"neighborhood.":[58],"believe":[60],"that":[61],"this":[62,83],"formulation":[63],"combines":[64],"the":[65,71,75,111],"regularizing":[66],"effect":[67],"dropout":[69],"smoothing":[72],"effects":[73],"convolution":[76],"operation.":[77],"In":[78],"addition":[79],"convolutional":[81],"paper":[84],"also":[85],"proposes":[86],"using":[87],"word-piece":[89],"segmentations":[90],"as":[91,125,127],"scheme":[95],"during":[96,110],"training,":[97],"inspired":[98],"results":[100],"neural":[102],"machine":[103],"translation.":[104],"adopt":[106],"both":[107],"these":[108],"methods":[109],"transformer-transducer":[114],"models,":[117],"show":[119],"consistent":[120],"WER":[121],"improvements":[122],"Librispeech":[124],"well":[126],"across":[128],"different":[129],"languages.":[130]},"counts_by_year":[{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}