{"id":"https://openalex.org/W2795138957","doi":"https://doi.org/10.21437/interspeech.2018-1910","title":"Self-Attentional Acoustic Models","display_name":"Self-Attentional Acoustic Models","publication_year":2018,"publication_date":"2018-08-28","ids":{"openalex":"https://openalex.org/W2795138957","doi":"https://doi.org/10.21437/interspeech.2018-1910","mag":"2795138957"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2018-1910","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2018-1910","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2018","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/1803.09519","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5024736091","display_name":"Matthias Sperber","orcid":null},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Matthias Sperber","raw_affiliation_strings":["Karlsruhe Institute of Technology, Karlsruhe, Germany"],"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology, Karlsruhe, Germany","institution_ids":["https://openalex.org/I102335020"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046084081","display_name":"Jan Niehues","orcid":"https://orcid.org/0000-0002-4231-6543"},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Jan Niehues","raw_affiliation_strings":["Karlsruhe Institute of Technology, Karlsruhe, Germany"],"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology, Karlsruhe, Germany","institution_ids":["https://openalex.org/I102335020"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068811427","display_name":"Graham Neubig","orcid":"https://orcid.org/0000-0002-2072-3789"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Graham Neubig","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, United States"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, United States","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065183855","display_name":"Sebastian St\u00fcker","orcid":null},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Sebastian St\u00fcker","raw_affiliation_strings":["Karlsruhe Institute of Technology, Karlsruhe, Germany"],"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology, Karlsruhe, Germany","institution_ids":["https://openalex.org/I102335020"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110453805","display_name":"Alex Waibel","orcid":null},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Alex Waibel","raw_affiliation_strings":["Karlsruhe Institute of Technology, Karlsruhe, Germany"],"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology, Karlsruhe, Germany","institution_ids":["https://openalex.org/I102335020"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5024736091"],"corresponding_institution_ids":["https://openalex.org/I102335020"],"apc_list":null,"apc_paid":null,"fwci":4.5561,"has_fulltext":true,"cited_by_count":29,"citation_normalized_percentile":{"value":0.95638825,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"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/T11309","display_name":"Music and Audio Processing","score":0.9997000098228455,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9980000257492065,"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/interpretability","display_name":"Interpretability","score":0.8059557676315308},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7618992328643799},{"id":"https://openalex.org/keywords/upsampling","display_name":"Upsampling","score":0.614517092704773},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5920114517211914},{"id":"https://openalex.org/keywords/pairwise-comparison","display_name":"Pairwise comparison","score":0.4568648040294647},{"id":"https://openalex.org/keywords/range","display_name":"Range (aeronautics)","score":0.4400981664657593},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.4282086491584778},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.34871596097946167}],"concepts":[{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.8059557676315308},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7618992328643799},{"id":"https://openalex.org/C110384440","wikidata":"https://www.wikidata.org/wiki/Q1143270","display_name":"Upsampling","level":3,"score":0.614517092704773},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5920114517211914},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.4568648040294647},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.4400981664657593},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.4282086491584778},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34871596097946167},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.0},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0},{"id":"https://openalex.org/C54355233","wikidata":"https://www.wikidata.org/wiki/Q7162","display_name":"Genetics","level":1,"score":0.0}],"mesh":[],"locations_count":5,"locations":[{"id":"doi:10.21437/interspeech.2018-1910","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2018-1910","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2018","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:1803.09519","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1803.09519","pdf_url":"https://arxiv.org/pdf/1803.09519","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":"","raw_type":"text"},{"id":"mag:2795138957","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/1803.09519","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.1803.09519","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1803.09519","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"},{"id":"doi:10.5445/ir/1000087581","is_oa":true,"landing_page_url":"https://doi.org/10.5445/ir/1000087581","pdf_url":null,"source":{"id":"https://openalex.org/S7407052948","display_name":"KITopen","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":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article-journal"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:1803.09519","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1803.09519","pdf_url":"https://arxiv.org/pdf/1803.09519","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":"","raw_type":"text"},"sustainable_development_goals":[{"score":0.699999988079071,"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2795138957.pdf","grobid_xml":"https://content.openalex.org/works/W2795138957.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W581956982","https://openalex.org/W1522301498","https://openalex.org/W1753482797","https://openalex.org/W2293634267","https://openalex.org/W2327501763","https://openalex.org/W2530876040","https://openalex.org/W2613904329","https://openalex.org/W2764124976","https://openalex.org/W2773734397","https://openalex.org/W2802023636","https://openalex.org/W2949888546","https://openalex.org/W2964308564"],"related_works":["https://openalex.org/W2963403868","https://openalex.org/W2892009249","https://openalex.org/W2802023636","https://openalex.org/W2064675550","https://openalex.org/W1524333225","https://openalex.org/W2936774411","https://openalex.org/W2964308564","https://openalex.org/W2963341956","https://openalex.org/W2962784628","https://openalex.org/W1902237438","https://openalex.org/W2972818416","https://openalex.org/W2911291251","https://openalex.org/W2964302946","https://openalex.org/W2798657914","https://openalex.org/W2407080277","https://openalex.org/W1494198834","https://openalex.org/W2514741789","https://openalex.org/W2327501763","https://openalex.org/W2101105183","https://openalex.org/W1522301498"],"abstract_inverted_index":{"Self-attention":[0],"is":[1,155],"a":[2,74,116,134,156,168],"method":[3],"of":[4,7,107,158,172],"encoding":[5],"sequences":[6],"vectors":[8,12],"by":[9],"relating":[10],"these":[11,59],"to":[13,34,36,40,51,57,83,99,103,147],"each-other":[14],"based":[15,137],"on":[16,138],"pairwise":[17],"similarities.":[18],"These":[19],"models":[20,98],"have":[21],"recently":[22],"shown":[23],"promising":[24],"results":[25],"for":[26],"modeling":[27,38,43],"discrete":[28],"sequences,":[29],"but":[30],"they":[31],"are":[32,90],"non-trivial":[33],"apply":[35,49],"acoustic":[37,52,112,160],"due":[39],"computational":[41],"and":[42,92,96,162],"issues.":[44],"In":[45],"this":[46,100],"paper,":[47],"we":[48,71,78,114,151],"self-attention":[50,62,165],"modeling,":[53],"proposing":[54],"several":[55],"improvements":[56],"mitigate":[58],"issues:":[60],"First,":[61],"memory":[63],"grows":[64],"quadratically":[65],"in":[66,110],"the":[67,88,105,111,125],"sequence":[68],"length,":[69],"which":[70],"address":[72],"through":[73],"downsampling":[75],"technique.":[76],"Second,":[77],"find":[79,129,152],"that":[80,120,130,153,164],"previous":[81],"approaches":[82,133],"incorporate":[84],"position":[85],"information":[86],"into":[87],"model":[89,132],"unsuitable":[91],"explore":[93],"other":[94],"representations":[95],"hybrid":[97],"end.":[101],"Third,":[102],"stress":[104],"importance":[106],"local":[108],"context":[109,126],"signal,":[113],"propose":[115],"Gaussian":[117],"biasing":[118],"approach":[119],"allows":[121],"explicit":[122],"control":[123],"over":[124],"range.":[127],"Experiments":[128],"our":[131],"strong":[135],"baseline":[136],"LSTMs":[139],"with":[140],"network-in-network":[141],"connections":[142],"while":[143],"being":[144],"much":[145],"faster":[146],"compute.":[148],"Besides":[149],"speed,":[150],"interpretability":[154],"strength":[157],"self-attentional":[159],"models,":[161],"demonstrate":[163],"heads":[166],"learn":[167],"linguistically":[169],"plausible":[170],"division":[171],"labor.":[173]},"counts_by_year":[{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":8},{"year":2020,"cited_by_count":11},{"year":2019,"cited_by_count":8}],"updated_date":"2026-03-10T16:38:18.471706","created_date":"2025-10-10T00:00:00"}