{"id":"https://openalex.org/W3005701363","doi":"https://doi.org/10.1109/icassp40776.2020.9053832","title":"Multitask Learning with Capsule Networks for Speech-to-Intent Applications","display_name":"Multitask Learning with Capsule Networks for Speech-to-Intent Applications","publication_year":2020,"publication_date":"2020-04-09","ids":{"openalex":"https://openalex.org/W3005701363","doi":"https://doi.org/10.1109/icassp40776.2020.9053832","mag":"3005701363"},"language":"en","primary_location":{"id":"doi:10.1109/icassp40776.2020.9053832","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp40776.2020.9053832","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2002.07450","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5009033650","display_name":"Jakob Poncelet","orcid":"https://orcid.org/0000-0002-5157-7577"},"institutions":[{"id":"https://openalex.org/I99464096","display_name":"KU Leuven","ror":"https://ror.org/05f950310","country_code":"BE","type":"education","lineage":["https://openalex.org/I99464096"]}],"countries":["BE"],"is_corresponding":true,"raw_author_name":"Jakob Poncelet","raw_affiliation_strings":["Department Electrical Engineering ESAT-PSI, KU Leuven, Leuven, Belgium","KU Leuven,Department Electrical Engineering ESAT-PSI,Leuven,Belgium,B-3001"],"affiliations":[{"raw_affiliation_string":"Department Electrical Engineering ESAT-PSI, KU Leuven, Leuven, Belgium","institution_ids":["https://openalex.org/I99464096"]},{"raw_affiliation_string":"KU Leuven,Department Electrical Engineering ESAT-PSI,Leuven,Belgium,B-3001","institution_ids":["https://openalex.org/I99464096"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5087514947","display_name":"Hugo Van hamme","orcid":"https://orcid.org/0000-0003-1331-5186"},"institutions":[{"id":"https://openalex.org/I99464096","display_name":"KU Leuven","ror":"https://ror.org/05f950310","country_code":"BE","type":"education","lineage":["https://openalex.org/I99464096"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Hugo Van hamme","raw_affiliation_strings":["Department Electrical Engineering ESAT-PSI, KU Leuven, Leuven, Belgium","Katholieke Universiteit Leuven"],"affiliations":[{"raw_affiliation_string":"Department Electrical Engineering ESAT-PSI, KU Leuven, Leuven, Belgium","institution_ids":["https://openalex.org/I99464096"]},{"raw_affiliation_string":"Katholieke Universiteit Leuven","institution_ids":["https://openalex.org/I99464096"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5009033650"],"corresponding_institution_ids":["https://openalex.org/I99464096"],"apc_list":null,"apc_paid":null,"fwci":0.14685955,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.51066401,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":"28","issue":null,"first_page":"8494","last_page":"8498"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","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"}},"topics":[{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","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"}},{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.9968000054359436,"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":0.9957000017166138,"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":0.8372483253479004},{"id":"https://openalex.org/keywords/utterance","display_name":"Utterance","score":0.7521864175796509},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6379835605621338},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6003828644752502},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5522046089172363},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.49895262718200684},{"id":"https://openalex.org/keywords/multi-task-learning","display_name":"Multi-task learning","score":0.4635787010192871},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4468532204627991},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.4333886504173279},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.34234005212783813}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8372483253479004},{"id":"https://openalex.org/C2775852435","wikidata":"https://www.wikidata.org/wiki/Q258403","display_name":"Utterance","level":2,"score":0.7521864175796509},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6379835605621338},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6003828644752502},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5522046089172363},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.49895262718200684},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.4635787010192871},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4468532204627991},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.4333886504173279},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.34234005212783813},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","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/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"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/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0}],"mesh":[],"locations_count":6,"locations":[{"id":"doi:10.1109/icassp40776.2020.9053832","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp40776.2020.9053832","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2002.07450","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2002.07450","pdf_url":"https://arxiv.org/pdf/2002.07450","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":"mag:3005701363","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2002.07450.pdf","pdf_url":null,"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":"arXiv (Cornell University)","raw_type":null},{"id":"pmh:oai:lirias2repo.kuleuven.be:123456789/650414","is_oa":true,"landing_page_url":"https://lirias.kuleuven.be/bitstream/123456789/650414/2/4464_postprint.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306401954","display_name":"Lirias (KU Leuven)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I99464096","host_organization_name":"KU Leuven","host_organization_lineage":["https://openalex.org/I99464096"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"45th IEEE international conference on acoustics, speech, and signal processing - ICASSP 2020, Barcelona, Spain ONLINE DUE TO COVID-19, 4-8 May 2020","raw_type":"info:eu-repo/semantics/publishedVersion"},{"id":"doi:10.48550/arxiv.2002.07450","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2002.07450","pdf_url":null,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article-journal"},{"id":"doi:10.17023/m1t3-pm13","is_oa":true,"landing_page_url":"https://doi.org/10.17023/m1t3-pm13","pdf_url":null,"source":{"id":"https://openalex.org/S7407051697","display_name":"IEEE RESOURCE CENTERS","issn_l":null,"issn":[],"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":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2002.07450","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2002.07450","pdf_url":"https://arxiv.org/pdf/2002.07450","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":[{"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions","score":0.7099999785423279}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W1989271290","https://openalex.org/W2092881279","https://openalex.org/W2157423596","https://openalex.org/W2401277329","https://openalex.org/W2891229414","https://openalex.org/W2894164357","https://openalex.org/W2913340405","https://openalex.org/W2963288440","https://openalex.org/W2963703618","https://openalex.org/W2964108264","https://openalex.org/W2964243145","https://openalex.org/W2972525948","https://openalex.org/W2972584841","https://openalex.org/W6743446608","https://openalex.org/W6755077353"],"related_works":["https://openalex.org/W3016219267","https://openalex.org/W3143995484","https://openalex.org/W319272071","https://openalex.org/W2612030847","https://openalex.org/W3107089030","https://openalex.org/W33124365","https://openalex.org/W64020921","https://openalex.org/W1551722583","https://openalex.org/W2141204697","https://openalex.org/W2741129224","https://openalex.org/W2741633070","https://openalex.org/W3036499953","https://openalex.org/W1584278479","https://openalex.org/W2901611695","https://openalex.org/W2740883964","https://openalex.org/W3212357841","https://openalex.org/W3012945836","https://openalex.org/W2869408871","https://openalex.org/W2133764590","https://openalex.org/W132247308"],"abstract_inverted_index":{"Voice":[0],"controlled":[1],"applications":[2],"can":[3,134,140],"be":[4,105,157],"a":[5,46,70,75,145,160,193,197],"great":[6],"aid":[7],"to":[8,19,68,89,94,116,123,162,171,196],"society,":[9],"especially":[10],"for":[11,52,58],"physically":[12],"challenged":[13],"people.":[14],"However":[15],"this":[16,128,188],"requires":[17],"robustness":[18],"all":[20],"kinds":[21],"of":[22,44,55,101,144,175],"variations":[23],"in":[24,48,166],"speech.":[25],"A":[26],"spoken":[27],"language":[28],"understanding":[29],"system":[30,47],"that":[31],"learns":[32,88,170],"from":[33,38,192],"interaction":[34],"with":[35,159],"and":[36,51,77],"demonstrations":[37],"the":[39,42,66,86,91,95,99,102,142,148,173,176,180,184],"user,":[40],"allows":[41],"use":[43],"such":[45],"different":[49,53],"settings":[50],"types":[54],"speech,":[56,62],"even":[57],"deviant":[59],"or":[60],"impaired":[61],"while":[63],"also":[64],"allowing":[65],"user":[67,73,103],"choose":[69],"phrasing.":[71],"The":[72,152],"gives":[74],"command":[76],"enters":[78],"its":[79,167],"intent":[80],"through":[81],"an":[82],"interface,":[83],"after":[84],"which":[85,138],"model":[87,146],"map":[90],"speech":[92],"directly":[93],"right":[96],"action.":[97],"Since":[98],"effort":[100],"should":[104],"as":[106,108],"low":[107],"possible,":[109],"capsule":[110,154,185],"networks":[111],"have":[112],"drawn":[113],"interest":[114],"due":[115],"potentially":[117],"needing":[118],"little":[119],"training":[120],"data":[121],"compared":[122],"deeper":[124],"neural":[125],"networks.":[126],"In":[127],"paper,":[129],"we":[130,190],"show":[131],"how":[132],"capsules":[133],"incorporate":[135],"multitask":[136],"learning,":[137],"often":[139],"improve":[141],"performance":[143],"when":[147],"task":[149],"is":[150],"difficult.":[151],"basic":[153],"network":[155],"will":[156],"expanded":[158],"regularisation":[161],"create":[163],"more":[164],"structure":[165],"output:":[168],"it":[169],"identify":[172],"speaker":[174,194,198],"utterance":[177],"by":[178],"forcing":[179],"required":[181],"information":[182],"into":[183],"vectors.":[186],"To":[187],"end":[189],"move":[191],"dependent":[195],"independent":[199],"setting.":[200]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2020,"cited_by_count":1}],"updated_date":"2026-02-09T09:26:11.010843","created_date":"2025-10-10T00:00:00"}