{"id":"https://openalex.org/W4372266841","doi":"https://doi.org/10.1109/icassp49357.2023.10095436","title":"Filterbank Learning for Noise-Robust Small-Footprint Keyword Spotting","display_name":"Filterbank Learning for Noise-Robust Small-Footprint Keyword Spotting","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372266841","doi":"https://doi.org/10.1109/icassp49357.2023.10095436"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10095436","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10095436","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2211.10565","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5048592470","display_name":"Iv\u00e1n L\u00f3pez\u2010Espejo","orcid":"https://orcid.org/0000-0001-8634-7897"},"institutions":[{"id":"https://openalex.org/I162577319","display_name":"The University of Texas at Dallas","ror":"https://ror.org/049emcs32","country_code":"US","type":"education","lineage":["https://openalex.org/I162577319"]},{"id":"https://openalex.org/I891191580","display_name":"Aalborg University","ror":"https://ror.org/04m5j1k67","country_code":"DK","type":"education","lineage":["https://openalex.org/I891191580"]}],"countries":["DK","US"],"is_corresponding":false,"raw_author_name":"Iv\u00e1n L\u00f3pez-Espejo","raw_affiliation_strings":["Aalborg University,Department of Electronic Systems,Denmark","Center for Robust Speech Systems (CRSS), The University of Texas at Dallas, USA","Department of Electronic Systems, Aalborg University, Denmark"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Aalborg University,Department of Electronic Systems,Denmark","institution_ids":["https://openalex.org/I891191580"]},{"raw_affiliation_string":"Center for Robust Speech Systems (CRSS), The University of Texas at Dallas, USA","institution_ids":["https://openalex.org/I162577319"]},{"raw_affiliation_string":"Department of Electronic Systems, Aalborg University, Denmark","institution_ids":["https://openalex.org/I891191580"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002916261","display_name":"Ram C. M. C. Shekar","orcid":null},"institutions":[{"id":"https://openalex.org/I162577319","display_name":"The University of Texas at Dallas","ror":"https://ror.org/049emcs32","country_code":"US","type":"education","lineage":["https://openalex.org/I162577319"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ram C. M. C. Shekar","raw_affiliation_strings":["The University of Texas at Dallas,Center for Robust Speech Systems (CRSS),USA","Center for Robust Speech Systems (CRSS), The University of Texas at Dallas, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Texas at Dallas,Center for Robust Speech Systems (CRSS),USA","institution_ids":["https://openalex.org/I162577319"]},{"raw_affiliation_string":"Center for Robust Speech Systems (CRSS), The University of Texas at Dallas, USA","institution_ids":["https://openalex.org/I162577319"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090108098","display_name":"Zheng\u2010Hua Tan","orcid":"https://orcid.org/0000-0001-6856-8928"},"institutions":[{"id":"https://openalex.org/I891191580","display_name":"Aalborg University","ror":"https://ror.org/04m5j1k67","country_code":"DK","type":"education","lineage":["https://openalex.org/I891191580"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Zheng-Hua Tan","raw_affiliation_strings":["Aalborg University,Department of Electronic Systems,Denmark","Department of Electronic Systems, Aalborg University, Denmark"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Aalborg University,Department of Electronic Systems,Denmark","institution_ids":["https://openalex.org/I891191580"]},{"raw_affiliation_string":"Department of Electronic Systems, Aalborg University, Denmark","institution_ids":["https://openalex.org/I891191580"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101679595","display_name":"Jesper Jensen","orcid":"https://orcid.org/0000-0003-1478-622X"},"institutions":[{"id":"https://openalex.org/I4210130492","display_name":"Oticon Medical (Denmark)","ror":"https://ror.org/03c5ds320","country_code":"DK","type":"company","lineage":["https://openalex.org/I4210130492"]},{"id":"https://openalex.org/I891191580","display_name":"Aalborg University","ror":"https://ror.org/04m5j1k67","country_code":"DK","type":"education","lineage":["https://openalex.org/I891191580"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Jesper Jensen","raw_affiliation_strings":["Aalborg University,Department of Electronic Systems,Denmark","Oticon A/S, Denmark","Department of Electronic Systems, Aalborg University, Denmark"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Aalborg University,Department of Electronic Systems,Denmark","institution_ids":["https://openalex.org/I891191580"]},{"raw_affiliation_string":"Oticon A/S, Denmark","institution_ids":["https://openalex.org/I4210130492"]},{"raw_affiliation_string":"Department of Electronic Systems, Aalborg University, Denmark","institution_ids":["https://openalex.org/I891191580"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5057910370","display_name":"John H. L. Hansen","orcid":"https://orcid.org/0000-0003-1382-9929"},"institutions":[{"id":"https://openalex.org/I162577319","display_name":"The University of Texas at Dallas","ror":"https://ror.org/049emcs32","country_code":"US","type":"education","lineage":["https://openalex.org/I162577319"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"John H. L. Hansen","raw_affiliation_strings":["The University of Texas at Dallas,Center for Robust Speech Systems (CRSS),USA","Center for Robust Speech Systems (CRSS), The University of Texas at Dallas, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Texas at Dallas,Center for Robust Speech Systems (CRSS),USA","institution_ids":["https://openalex.org/I162577319"]},{"raw_affiliation_string":"Center for Robust Speech Systems (CRSS), The University of Texas at Dallas, USA","institution_ids":["https://openalex.org/I162577319"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.6473,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":{"value":0.73775416,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"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/T10860","display_name":"Speech and Audio Processing","score":0.9994000196456909,"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/T11309","display_name":"Music and Audio Processing","score":0.9994000196456909,"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/keyword-spotting","display_name":"Keyword spotting","score":0.856359601020813},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8151097893714905},{"id":"https://openalex.org/keywords/filter-bank","display_name":"Filter bank","score":0.7510541677474976},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6276432275772095},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6053929328918457},{"id":"https://openalex.org/keywords/energy-consumption","display_name":"Energy consumption","score":0.4586314260959625},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4420269727706909},{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.43617168068885803},{"id":"https://openalex.org/keywords/filter","display_name":"Filter (signal processing)","score":0.34819915890693665},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.12345990538597107},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08701157569885254}],"concepts":[{"id":"https://openalex.org/C2781213101","wikidata":"https://www.wikidata.org/wiki/Q6398558","display_name":"Keyword spotting","level":2,"score":0.856359601020813},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8151097893714905},{"id":"https://openalex.org/C100515483","wikidata":"https://www.wikidata.org/wiki/Q3268235","display_name":"Filter bank","level":3,"score":0.7510541677474976},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6276432275772095},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6053929328918457},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.4586314260959625},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4420269727706909},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.43617168068885803},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.34819915890693665},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.12345990538597107},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08701157569885254},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","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/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp49357.2023.10095436","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10095436","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},{"id":"pmh:oai:pure.atira.dk:publications/89842936-8df4-45d0-b12d-d539fbbf483e","is_oa":true,"landing_page_url":"https://vbn.aau.dk/da/publications/89842936-8df4-45d0-b12d-d539fbbf483e","pdf_url":"https://arxiv.org/pdf/2211.10565","source":{"id":"https://openalex.org/S4306401731","display_name":"VBN Forskningsportal (Aalborg Universitet)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I891191580","host_organization_name":"Aalborg University","host_organization_lineage":["https://openalex.org/I891191580"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Espejo , I L , Shekar , R C M C , Tan , Z-H , Jensen , J & Hansen , J 2023 , Filterbank Learning for Noise-Robust Small-Footprint Keyword Spotting . in ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing, Proceedings . , 10095436 , IEEE , International Conference on Acoustics Speech and Signal Processing (ICASSP) , 48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023 , Rhodes Island , Greece , 04/06/2023 . https://doi.org/10.1109/ICASSP49357.2023.10095436","raw_type":"contributionToPeriodical"}],"best_oa_location":{"id":"pmh:oai:pure.atira.dk:publications/89842936-8df4-45d0-b12d-d539fbbf483e","is_oa":true,"landing_page_url":"https://vbn.aau.dk/da/publications/89842936-8df4-45d0-b12d-d539fbbf483e","pdf_url":"https://arxiv.org/pdf/2211.10565","source":{"id":"https://openalex.org/S4306401731","display_name":"VBN Forskningsportal (Aalborg Universitet)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I891191580","host_organization_name":"Aalborg University","host_organization_lineage":["https://openalex.org/I891191580"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Espejo , I L , Shekar , R C M C , Tan , Z-H , Jensen , J & Hansen , J 2023 , Filterbank Learning for Noise-Robust Small-Footprint Keyword Spotting . in ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing, Proceedings . , 10095436 , IEEE , International Conference on Acoustics Speech and Signal Processing (ICASSP) , 48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023 , Rhodes Island , Greece , 04/06/2023 . https://doi.org/10.1109/ICASSP49357.2023.10095436","raw_type":"contributionToPeriodical"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy","score":0.8999999761581421}],"awards":[{"id":"https://openalex.org/G2717016369","display_name":"A Giant Leap for Keyword Spotting","funder_award_id":"101062614","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"}],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4372266841.pdf","grobid_xml":"https://content.openalex.org/works/W4372266841.grobid-xml"},"referenced_works_count":17,"referenced_works":["https://openalex.org/W1974387177","https://openalex.org/W2095705004","https://openalex.org/W2559809918","https://openalex.org/W2797583228","https://openalex.org/W2964052309","https://openalex.org/W2964187693","https://openalex.org/W2964250984","https://openalex.org/W2972869267","https://openalex.org/W3015287265","https://openalex.org/W3114848746","https://openalex.org/W3154934206","https://openalex.org/W3176579201","https://openalex.org/W3196496149","https://openalex.org/W4206567542","https://openalex.org/W4293363567","https://openalex.org/W6674330103","https://openalex.org/W6750665317"],"related_works":["https://openalex.org/W2114097550","https://openalex.org/W4385352507","https://openalex.org/W2918559346","https://openalex.org/W84309476","https://openalex.org/W4286904253","https://openalex.org/W3114848746","https://openalex.org/W3031469774","https://openalex.org/W4287773882","https://openalex.org/W2065885792","https://openalex.org/W2159300429"],"abstract_inverted_index":{"In":[0,22],"the":[1,7,37,46,81],"context":[2],"of":[3,9,39,48,80,99,127],"keyword":[4],"spotting":[5],"(KWS),":[6],"replacement":[8],"handcrafted":[10,31],"speech":[11,32],"features":[12,15,33,115,119],"by":[13],"learnable":[14],"has":[16],"not":[17],"yielded":[18],"superior":[19],"KWS":[20,35,53,70,124],"performance.":[21],"this":[23],"study,":[24],"we":[25],"demonstrate":[26],"that":[27,87],"filterbank":[28,40,88],"learning":[29,89],"outperforms":[30],"for":[34],"whenever":[36],"number":[38,47],"channels":[41,49],"is":[42,64,106],"severely":[43],"decreased.":[44],"Reducing":[45],"might":[50],"yield":[51],"certain":[52],"performance":[54],"drop,":[55],"but":[56],"also":[57],"a":[58,77,96,122,133],"substantial":[59],"energy":[60,135],"consumption":[61,136],"reduction,":[62],"which":[63],"key":[65],"when":[66,104],"deploying":[67],"common":[68],"always-on":[69],"on":[71,76],"low-resource":[72],"devices.":[73],"Experimental":[74],"results":[75],"noisy":[78],"version":[79],"Google":[82],"Speech":[83],"Commands":[84],"Dataset":[85],"show":[86],"adapts":[90],"to":[91,94,101,116,121],"noise":[92],"characteristics":[93],"provide":[95],"higher":[97],"degree":[98],"robustness":[100],"noise,":[102],"especially":[103],"dropout":[105],"integrated.":[107],"Thus,":[108],"switching":[109],"from":[110],"typically":[111],"used":[112],"40-channel":[113],"log-Mel":[114],"8channel":[117],"learned":[118],"leads":[120],"relative":[123],"accuracy":[125],"loss":[126],"only":[128],"3.5%":[129],"while":[130],"simultaneously":[131],"achieving":[132],"6.3\u00d7":[134],"reduction.":[137]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2}],"updated_date":"2026-06-22T08:00:12.763002","created_date":"2025-10-10T00:00:00"}