{"id":"https://openalex.org/W3174234851","doi":"https://doi.org/10.21437/interspeech.2021-1110","title":"StableEmit: Selection Probability Discount for Reducing Emission Latency of Streaming Monotonic Attention ASR","display_name":"StableEmit: Selection Probability Discount for Reducing Emission Latency of Streaming Monotonic Attention ASR","publication_year":2021,"publication_date":"2021-08-27","ids":{"openalex":"https://openalex.org/W3174234851","doi":"https://doi.org/10.21437/interspeech.2021-1110","mag":"3174234851"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2021-1110","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-1110","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2021","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2107.00635","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5040282669","display_name":"Hirofumi Inaguma","orcid":"https://orcid.org/0000-0003-0610-1251"},"institutions":[{"id":"https://openalex.org/I22299242","display_name":"Kyoto University","ror":"https://ror.org/02kpeqv85","country_code":"JP","type":"education","lineage":["https://openalex.org/I22299242"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Hirofumi Inaguma","raw_affiliation_strings":["Kyoto University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kyoto University","institution_ids":["https://openalex.org/I22299242"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5038044080","display_name":"Tatsuya Kawahara","orcid":"https://orcid.org/0000-0002-2686-2296"},"institutions":[{"id":"https://openalex.org/I22299242","display_name":"Kyoto University","ror":"https://ror.org/02kpeqv85","country_code":"JP","type":"education","lineage":["https://openalex.org/I22299242"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Tatsuya Kawahara","raw_affiliation_strings":["Kyoto University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kyoto University","institution_ids":["https://openalex.org/I22299242"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I22299242"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1817","last_page":"1821"},"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":0.9994999766349792,"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/T10860","display_name":"Speech and Audio Processing","score":0.9988999962806702,"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/latency","display_name":"Latency (audio)","score":0.7456838488578796},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.7421673536300659},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7304393649101257},{"id":"https://openalex.org/keywords/monotonic-function","display_name":"Monotonic function","score":0.7180693745613098},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5773019790649414},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.5181171298027039},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5031694769859314},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4051244258880615},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.37666454911231995},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.12100759148597717},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.12061381340026855}],"concepts":[{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.7456838488578796},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7421673536300659},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7304393649101257},{"id":"https://openalex.org/C72169020","wikidata":"https://www.wikidata.org/wiki/Q194404","display_name":"Monotonic function","level":2,"score":0.7180693745613098},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5773019790649414},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.5181171298027039},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5031694769859314},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4051244258880615},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.37666454911231995},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.12100759148597717},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.12061381340026855},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.21437/interspeech.2021-1110","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-1110","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2021","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2107.00635","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2107.00635","pdf_url":"https://arxiv.org/pdf/2107.00635","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":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"mag:3174234851","is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/2107.00635","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.2107.00635","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2107.00635","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":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2107.00635","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2107.00635","pdf_url":"https://arxiv.org/pdf/2107.00635","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":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3174234851.pdf","grobid_xml":"https://content.openalex.org/works/W3174234851.grobid-xml"},"referenced_works_count":28,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1522301498","https://openalex.org/W1524333225","https://openalex.org/W1828163288","https://openalex.org/W1836465849","https://openalex.org/W2127141656","https://openalex.org/W2250357346","https://openalex.org/W2291513470","https://openalex.org/W2407080277","https://openalex.org/W2577366047","https://openalex.org/W2605141709","https://openalex.org/W2795783309","https://openalex.org/W2913718171","https://openalex.org/W2936774411","https://openalex.org/W2962784628","https://openalex.org/W2963382396","https://openalex.org/W2963925437","https://openalex.org/W2964272710","https://openalex.org/W2972977747","https://openalex.org/W3015315932","https://openalex.org/W3015654466","https://openalex.org/W3018242354","https://openalex.org/W3094759821","https://openalex.org/W3097777922","https://openalex.org/W3128478537","https://openalex.org/W3133822476","https://openalex.org/W3149509723","https://openalex.org/W3163213339"],"related_works":["https://openalex.org/W3163213339","https://openalex.org/W2139034093","https://openalex.org/W3212328129","https://openalex.org/W3026505846","https://openalex.org/W2566478918","https://openalex.org/W3117464205","https://openalex.org/W2340721073","https://openalex.org/W2197110155","https://openalex.org/W3205764742","https://openalex.org/W2979167137","https://openalex.org/W3115734461","https://openalex.org/W3118895645","https://openalex.org/W2769644379","https://openalex.org/W2913913178","https://openalex.org/W2962761403","https://openalex.org/W2780033725","https://openalex.org/W2890152612","https://openalex.org/W2740515525","https://openalex.org/W2750919002","https://openalex.org/W2951791189"],"abstract_inverted_index":{"While":[0],"attention-based":[1],"encoder-decoder":[2],"(AED)":[3],"models":[4,26],"have":[5,28],"been":[6],"successfully":[7],"extended":[8],"to":[9,50,54,70,73,97,128,147],"the":[10,25,36,52,79,99,108,111,117,150,166,170,178],"online":[11],"variants":[12],"for":[13,86,123],"streaming":[14],"automatic":[15],"speech":[16],"recognition":[17,167],"(ASR),":[18],"such":[19],"as":[20],"monotonic":[21,84],"chunkwise":[22],"attention":[23,85,101],"(MoChA),":[24],"still":[27],"a":[29,64,91,106,121,129],"large":[30],"label":[31],"emission":[32,125,132,171],"latency":[33,133,172],"because":[34],"of":[35,110,131,180],"unconstrained":[37],"end-to-end":[38],"training":[39],"objective.":[40],"Previous":[41],"works":[42],"tackled":[43],"this":[44,60],"problem":[45],"by":[46,90],"leveraging":[47],"alignment":[48,181],"information":[49,182],"control":[51],"timing":[53],"emit":[55,74],"tokens":[56,75],"during":[57,103],"training.":[58,104],"In":[59],"work,":[61],"we":[62],"propose":[63],"simple":[65],"alignment-free":[66],"regularization":[67],"method,":[68],"StableEmit,":[69],"encourage":[71],"MoChA":[72],"earlier.":[76],"StableEmit":[77,138,163],"discounts":[78],"selection":[80,112],"probabilities":[81,113],"in":[82,185],"hard":[83],"token":[87,124],"boundary":[88],"detection":[89],"constant":[92],"factor":[93],"and":[94,116,134,152,158,169],"regularizes":[95],"them":[96],"recover":[98],"total":[100],"mass":[102],"As":[105],"result,":[107],"scale":[109],"is":[114,183],"increased,":[115],"values":[118],"can":[119,139],"reach":[120],"threshold":[122],"earlier,":[126],"leading":[127],"reduction":[130],"deletion":[135],"errors.":[136],"Moreover,":[137],"be":[140],"combined":[141],"with":[142,156],"methods":[143],"that":[144,162,177],"constraint":[145],"alignments":[146],"further":[148],"improve":[149],"accuracy":[151],"latency.":[153],"Experimental":[154],"evaluations":[155],"LSTM":[157],"Conformer":[159],"encoders":[160],"demonstrate":[161],"significantly":[164],"reduces":[165],"errors":[168],"simultaneously.":[173],"We":[174],"also":[175],"show":[176],"use":[179],"complementary":[184],"both":[186],"metrics.":[187]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}