{"id":"https://openalex.org/W3132582893","doi":"https://doi.org/10.48550/arxiv.2103.04792","title":"An Ultra-low Power RNN Classifier for Always-On Voice Wake-Up Detection Robust to Real-World Scenarios","display_name":"An Ultra-low Power RNN Classifier for Always-On Voice Wake-Up Detection Robust to Real-World Scenarios","publication_year":2021,"publication_date":"2021-03-08","ids":{"openalex":"https://openalex.org/W3132582893","doi":"https://doi.org/10.48550/arxiv.2103.04792","mag":"3132582893"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2103.04792","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2103.04792","pdf_url":"https://arxiv.org/pdf/2103.04792","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2103.04792","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5010860677","display_name":"Emmanuel Hardy","orcid":"https://orcid.org/0000-0002-9208-1674","observed_orcids":["https://orcid.org/0000-0002-9208-1674"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hardy, Emmanuel","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5031744532","display_name":"Franck Badets","orcid":"https://orcid.org/0000-0002-8331-4042","observed_orcids":["https://orcid.org/0000-0002-8331-4042"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Badets, Franck","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"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":0.3698999881744385,"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":0.3698999881744385,"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/T10363","display_name":"Low-power high-performance VLSI design","score":0.3043999969959259,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10323","display_name":"Analog and Mixed-Signal Circuit Design","score":0.11469999700784683,"subfield":{"id":"https://openalex.org/subfields/2204","display_name":"Biomedical Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/ultra-low-power","display_name":"ultra-low power","score":0.7459999918937683},{"id":"https://openalex.org/keywords/recurrent-neural-networks","display_name":"recurrent neural networks","score":0.9350000023841858},{"id":"https://openalex.org/keywords/noise-robustness","display_name":"noise robustness","score":0.7170000076293945},{"id":"https://openalex.org/keywords/loss-function-design","display_name":"loss function design","score":0.652999997138977},{"id":"https://openalex.org/keywords/duty-cycle","display_name":"duty cycle","score":0.6859999895095825},{"id":"https://openalex.org/keywords/65-nm-cmos","display_name":"65 nm CMOS","score":0.75},{"id":"https://openalex.org/keywords/memory-usage","display_name":"memory usage","score":0.734000027179718},{"id":"https://openalex.org/keywords/keyword-spotting","display_name":"keyword spotting","score":0.6650000214576721},{"id":"https://openalex.org/keywords/voice-activity-detection","display_name":"voice activity detection","score":0.7149999737739563}],"concepts":[{"id":"https://openalex.org/C2781213101","wikidata":"https://www.wikidata.org/wiki/Q6398558","display_name":"Keyword spotting","level":2,"score":0.8059653043746948},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7382106184959412},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.6987732648849487},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6725847721099854},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.6326215267181396},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.589137077331543},{"id":"https://openalex.org/C100675267","wikidata":"https://www.wikidata.org/wiki/Q1371624","display_name":"Background noise","level":2,"score":0.43153178691864014},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39705681800842285},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.2966088652610779},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.217749685049057},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.12044063210487366}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:2103.04792","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2103.04792","pdf_url":"https://arxiv.org/pdf/2103.04792","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2103.04792","doi":"https://doi.org/10.48550/arxiv.2103.04792","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2103.04792","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":false,"raw_source_name":null,"raw_type":"Preprint"},{"id":"mag:3132582893","doi":null,"is_oa":false,"landing_page_url":"https://openreview.net/pdf?id=_Qtnx5BY6ZT","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":null}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2103.04792","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2103.04792","pdf_url":"https://arxiv.org/pdf/2103.04792","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W1924770834","https://openalex.org/W1999454387","https://openalex.org/W2048708674","https://openalex.org/W2064675550","https://openalex.org/W2115717467","https://openalex.org/W2157331557","https://openalex.org/W2219249508","https://openalex.org/W2319453305","https://openalex.org/W2512629640","https://openalex.org/W2587529061","https://openalex.org/W2602634800","https://openalex.org/W2767895109","https://openalex.org/W2797583228","https://openalex.org/W2889847962","https://openalex.org/W2930364467","https://openalex.org/W2938874599","https://openalex.org/W2944216546","https://openalex.org/W2953219395","https://openalex.org/W2973127906","https://openalex.org/W2997481719","https://openalex.org/W3012561096","https://openalex.org/W3158481788"],"related_works":["https://openalex.org/W7220126129","https://openalex.org/W2902572630","https://openalex.org/W2791524226","https://openalex.org/W2652238127","https://openalex.org/W2973127906","https://openalex.org/W4412191023","https://openalex.org/W2048497537","https://openalex.org/W3089633658","https://openalex.org/W3005319366","https://openalex.org/W6929431737"],"abstract_inverted_index":{"We":[0,97,118,146],"present":[1],"in":[2,46,134,173],"this":[3,153],"paper":[4],"an":[5,16,143,162],"ultra-low":[6],"power":[7,44,164],"(ULP)":[8],"Recurrent":[9],"Neural":[10],"Network":[11],"(RNN)":[12],"based":[13],"classifier":[14],"for":[15,25,81,127,169],"always-on":[17,50],"voice":[18],"Wake-Up":[19],"Sensor":[20],"(WUS)":[21],"with":[22,161],"performances":[23],"suitable":[24],"real-world":[26,79],"applications.":[27],"The":[28],"purpose":[29],"of":[30,49,84,142,150,166,181],"our":[31,71],"sensor":[32,73],"is":[33],"to":[34,74,77,102,156],"bring":[35],"down":[36],"by":[37,88],"at":[38],"least":[39],"a":[40,111,128,140,177],"factor":[41],"100":[42],"the":[43,65,91,94,105,148,157,170],"consumption":[45,165],"background":[47,136],"noise":[48,87],"speech":[51,85,106],"processing":[52],"algorithms":[53],"such":[54],"as":[55],"Automatic":[56],"Speech":[57],"Recognition,":[58],"Keyword":[59],"Spotting,":[60],"Speaker":[61],"Verification,":[62],"etc.":[63],"Unlike":[64],"other":[66,158],"published":[67],"approaches,":[68,160],"we":[69],"designed":[70],"wake-up":[72],"be":[75],"robust":[76],"unseen":[78],"noises":[80,137],"realistic":[82],"levels":[83],"and":[86,93,176],"carefully":[89],"designing":[90],"dataset":[92],"loss":[95],"function.":[96],"also":[98],"specifically":[99],"trained":[100],"it":[101],"mark":[103],"only":[104],"start":[107],"rather":[108],"than":[109,121,132],"adopting":[110],"traditional":[112],"Voice":[113],"Activity":[114],"Detection":[115],"(VAD)":[116],"approach.":[117],"achieve":[119],"less":[120,131],"3%":[122],"No":[123],"Trigger":[124],"Rate":[125],"(NTR)":[126],"duty":[129],"cycle":[130],"1%":[133],"challenging":[135],"pooled":[138],"using":[139],"model":[141],"analogue":[144],"front-end.":[145],"demonstrate":[147],"superiority":[149],"RNNs":[151],"on":[152],"task":[154],"compared":[155],"tested":[159],"estimated":[163],"45":[167],"nW":[168],"RNN":[171],"itself":[172],"65nm":[174],"CMOS":[175],"minimal":[178],"memory":[179],"footprint":[180],"0.52":[182],"kB.":[183]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":2}],"updated_date":"2026-11-03T10:50:19.383937","created_date":"2021-03-01T00:00:00"}