{"id":"https://openalex.org/W2981818381","doi":"https://doi.org/10.1109/icassp40776.2020.9054314","title":"End-to-End Architectures for ASR-Free Spoken Language Understanding","display_name":"End-to-End Architectures for ASR-Free Spoken Language Understanding","publication_year":2020,"publication_date":"2020-04-09","ids":{"openalex":"https://openalex.org/W2981818381","doi":"https://doi.org/10.1109/icassp40776.2020.9054314","mag":"2981818381"},"language":"en","primary_location":{"id":"doi:10.1109/icassp40776.2020.9054314","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp40776.2020.9054314","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/1910.10599","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5013653339","display_name":"Elisavet Palogiannidi","orcid":"https://orcid.org/0000-0003-0987-4049"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Elisavet Palogiannidi","raw_affiliation_strings":["Omilia - Conversational Intelligence, Athens, Greece","Omilia - Conversational Intelligence,Athens,Greece"],"affiliations":[{"raw_affiliation_string":"Omilia - Conversational Intelligence, Athens, Greece","institution_ids":[]},{"raw_affiliation_string":"Omilia - Conversational Intelligence,Athens,Greece","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046131361","display_name":"Ioannis Gkinis","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ioannis Gkinis","raw_affiliation_strings":["Omilia - Conversational Intelligence, Athens, Greece","Omilia - Conversational Intelligence,Athens,Greece"],"affiliations":[{"raw_affiliation_string":"Omilia - Conversational Intelligence, Athens, Greece","institution_ids":[]},{"raw_affiliation_string":"Omilia - Conversational Intelligence,Athens,Greece","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040232806","display_name":"George Mastrapas","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"George Mastrapas","raw_affiliation_strings":["Omilia - Conversational Intelligence, Athens, Greece","Omilia - Conversational Intelligence,Athens,Greece"],"affiliations":[{"raw_affiliation_string":"Omilia - Conversational Intelligence, Athens, Greece","institution_ids":[]},{"raw_affiliation_string":"Omilia - Conversational Intelligence,Athens,Greece","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040000490","display_name":"Petr Mizera","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Petr Mizera","raw_affiliation_strings":["Omilia - Conversational Intelligence, Athens, Greece","Omilia - Conversational Intelligence,Athens,Greece"],"affiliations":[{"raw_affiliation_string":"Omilia - Conversational Intelligence, Athens, Greece","institution_ids":[]},{"raw_affiliation_string":"Omilia - Conversational Intelligence,Athens,Greece","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5061939508","display_name":"Themos Stafylakis","orcid":"https://orcid.org/0000-0002-9227-3588"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Themos Stafylakis","raw_affiliation_strings":["Omilia - Conversational Intelligence, Athens, Greece","Omilia - Conversational Intelligence,Athens,Greece"],"affiliations":[{"raw_affiliation_string":"Omilia - Conversational Intelligence, Athens, Greece","institution_ids":[]},{"raw_affiliation_string":"Omilia - Conversational Intelligence,Athens,Greece","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5013653339"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.44057865,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":{"value":0.66792325,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"7974","last_page":"7978"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","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/T10028","display_name":"Topic Modeling","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":0.9998999834060669,"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/T12031","display_name":"Speech and dialogue systems","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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8259626626968384},{"id":"https://openalex.org/keywords/generalizability-theory","display_name":"Generalizability theory","score":0.7930462956428528},{"id":"https://openalex.org/keywords/natural-language-understanding","display_name":"Natural language understanding","score":0.7462559938430786},{"id":"https://openalex.org/keywords/spoken-language","display_name":"Spoken language","score":0.6826175451278687},{"id":"https://openalex.org/keywords/meaning","display_name":"Meaning (existential)","score":0.6120542883872986},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.6036325097084045},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5682730674743652},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5296413898468018},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5243630409240723},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.4364953637123108},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.3897842466831207},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.11955270171165466}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8259626626968384},{"id":"https://openalex.org/C27158222","wikidata":"https://www.wikidata.org/wiki/Q5532422","display_name":"Generalizability theory","level":2,"score":0.7930462956428528},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.7462559938430786},{"id":"https://openalex.org/C2776230583","wikidata":"https://www.wikidata.org/wiki/Q1322198","display_name":"Spoken language","level":2,"score":0.6826175451278687},{"id":"https://openalex.org/C2780876879","wikidata":"https://www.wikidata.org/wiki/Q3054749","display_name":"Meaning (existential)","level":2,"score":0.6120542883872986},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.6036325097084045},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5682730674743652},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5296413898468018},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5243630409240723},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.4364953637123108},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3897842466831207},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.11955270171165466},{"id":"https://openalex.org/C138496976","wikidata":"https://www.wikidata.org/wiki/Q175002","display_name":"Developmental psychology","level":1,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C542102704","wikidata":"https://www.wikidata.org/wiki/Q183257","display_name":"Psychotherapist","level":1,"score":0.0}],"mesh":[],"locations_count":5,"locations":[{"id":"doi:10.1109/icassp40776.2020.9054314","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp40776.2020.9054314","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:1910.10599","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1910.10599","pdf_url":"https://arxiv.org/pdf/1910.10599","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":null},{"id":"mag:2981818381","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/1910.10599.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":"doi:10.48550/arxiv.1910.10599","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1910.10599","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"},{"id":"doi:10.17023/0pvw-bb94","is_oa":true,"landing_page_url":"https://doi.org/10.17023/0pvw-bb94","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:1910.10599","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1910.10599","pdf_url":"https://arxiv.org/pdf/1910.10599","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":null},"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.8100000023841858}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2981818381.pdf","grobid_xml":"https://content.openalex.org/works/W2981818381.grobid-xml"},"referenced_works_count":33,"referenced_works":["https://openalex.org/W648786980","https://openalex.org/W648947103","https://openalex.org/W1524333225","https://openalex.org/W1649407914","https://openalex.org/W2007261869","https://openalex.org/W2099877156","https://openalex.org/W2219249508","https://openalex.org/W2250539671","https://openalex.org/W2514741789","https://openalex.org/W2736272491","https://openalex.org/W2786839803","https://openalex.org/W2803392141","https://openalex.org/W2891229414","https://openalex.org/W2901336235","https://openalex.org/W2905730339","https://openalex.org/W2928075308","https://openalex.org/W2933022734","https://openalex.org/W2963288440","https://openalex.org/W2963341956","https://openalex.org/W2965373594","https://openalex.org/W2972525948","https://openalex.org/W2972584841","https://openalex.org/W2978017171","https://openalex.org/W6621543089","https://openalex.org/W6631362777","https://openalex.org/W6674796821","https://openalex.org/W6688816777","https://openalex.org/W6748263443","https://openalex.org/W6755077353","https://openalex.org/W6755207826","https://openalex.org/W6757078890","https://openalex.org/W6766673545","https://openalex.org/W6768851824"],"related_works":["https://openalex.org/W3016262400","https://openalex.org/W3174008653","https://openalex.org/W3138431698","https://openalex.org/W2786839803","https://openalex.org/W2981486861","https://openalex.org/W3107401990","https://openalex.org/W2913960647","https://openalex.org/W3132012095","https://openalex.org/W3108858638","https://openalex.org/W3029914199","https://openalex.org/W2618015847","https://openalex.org/W3005352739","https://openalex.org/W3162313915","https://openalex.org/W2963230678","https://openalex.org/W2007261869","https://openalex.org/W3015747801","https://openalex.org/W3016042119","https://openalex.org/W3153839229","https://openalex.org/W3004225838","https://openalex.org/W3015267417"],"abstract_inverted_index":{"Spoken":[0],"Language":[1,40],"Understanding":[2,41],"(SLU)":[3],"is":[4,15,29],"the":[5,9,48,59,94,152],"problem":[6],"of":[7,73,86,108],"extracting":[8],"meaning":[10,46],"from":[11,47],"speech":[12,33],"utterances.":[13],"It":[14],"typically":[16],"addressed":[17],"as":[18,106],"a":[19,38,64,84],"two-step":[20],"problem,":[21],"where":[22,102],"an":[23],"Automatic":[24],"Speech":[25,98],"Recognition":[26],"(ASR)":[27],"model":[28,43,153],"employed":[30],"to":[31,44,93,145],"convert":[32],"into":[34,63],"text,":[35],"followed":[36],"by":[37,118],"Natural":[39],"(NLU)":[42],"extract":[45],"decoded":[49],"text.":[50],"Recently,":[51],"end-to-end":[52],"approaches":[53],"were":[54],"emerged,":[55],"aiming":[56],"at":[57],"unifying":[58],"ASR":[60,74,138],"and":[61,75,113,148],"NLU":[62],"single":[65],"SLU":[66],"deep":[67,120],"neural":[68],"architecture,":[69],"trained":[70],"using":[71,133],"combinations":[72,107],"NLU-level":[76],"recognition":[77],"units.":[78],"In":[79],"this":[80],"paper,":[81],"we":[82,149],"explore":[83],"set":[85],"recurrent":[87,121],"architectures":[88,122],"for":[89],"intent":[90],"classification,":[91],"tailored":[92],"recently":[95],"introduced":[96],"Fluent":[97],"Commands":[99],"(FSC)":[100],"dataset,":[101],"intents":[103],"are":[104],"formed":[105],"three":[109],"slots":[110],"(action,":[111],"object,":[112],"location).":[114],"We":[115,140],"show":[116,150],"that":[117,151],"combining":[119],"with":[123],"standard":[124],"data":[125],"augmentation,":[126],"state-of-the-art":[127],"results":[128],"can":[129,154],"be":[130],"attained,":[131],"without":[132],"ASR-level":[134],"targets":[135],"or":[136],"pretrained":[137],"models.":[139],"also":[141],"investigate":[142],"its":[143],"generalizability":[144],"new":[146],"wordings,":[147],"perform":[155],"reasonably":[156],"well":[157],"on":[158],"wordings":[159],"unseen":[160],"during":[161],"training.":[162]},"counts_by_year":[{"year":2020,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}