{"id":"https://openalex.org/W4308521107","doi":"https://doi.org/10.21437/iberspeech.2022-27","title":"An Experimental Study on Light Speech Features for Small-Footprint Keyword Spotting","display_name":"An Experimental Study on Light Speech Features for Small-Footprint Keyword Spotting","publication_year":2022,"publication_date":"2022-11-08","ids":{"openalex":"https://openalex.org/W4308521107","doi":"https://doi.org/10.21437/iberspeech.2022-27"},"language":"en","primary_location":{"id":"doi:10.21437/iberspeech.2022-27","is_oa":false,"landing_page_url":"https://doi.org/10.21437/iberspeech.2022-27","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IberSPEECH 2022","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://vbn.aau.dk/da/publications/9578dde1-bec6-40b5-8efd-ee47458bb2db","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/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":"Iv\u00e1n L\u00f3pez-Espejo","raw_affiliation_strings":["Department of Electronic Systems, Aalborg University, Denmark"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Systems, Aalborg University, Denmark","institution_ids":["https://openalex.org/I891191580"]}]},{"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":["Department of Electronic Systems, Aalborg University, Denmark"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Systems, Aalborg University, Denmark","institution_ids":["https://openalex.org/I891191580"]}]},{"author_position":"last","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":["Oticon A/S, Denmark","Department of Electronic Systems, Aalborg University, Denmark"],"raw_orcid":null,"affiliations":[{"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"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.1387,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.55925215,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":94},"biblio":{"volume":null,"issue":null,"first_page":"131","last_page":"135"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998000264167786,"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":0.9998000264167786,"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":0.9995999932289124,"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":0.9994999766349792,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/keyword-spotting","display_name":"Keyword spotting","score":0.9476271867752075},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7324655055999756},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.5938360691070557},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.5748644471168518},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.558601438999176},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.5336469411849976},{"id":"https://openalex.org/keywords/mel-frequency-cepstrum","display_name":"Mel-frequency cepstrum","score":0.5109166502952576},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.4874322712421417},{"id":"https://openalex.org/keywords/energy-consumption","display_name":"Energy consumption","score":0.4744452238082886},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4414827525615692},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.41852259635925293},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3856751620769501},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3709510564804077},{"id":"https://openalex.org/keywords/feature-extraction","display_name":"Feature extraction","score":0.3694359362125397},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1675669252872467},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08066973090171814}],"concepts":[{"id":"https://openalex.org/C2781213101","wikidata":"https://www.wikidata.org/wiki/Q6398558","display_name":"Keyword spotting","level":2,"score":0.9476271867752075},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7324655055999756},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.5938360691070557},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5748644471168518},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.558601438999176},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.5336469411849976},{"id":"https://openalex.org/C151989614","wikidata":"https://www.wikidata.org/wiki/Q440370","display_name":"Mel-frequency cepstrum","level":3,"score":0.5109166502952576},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.4874322712421417},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.4744452238082886},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4414827525615692},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.41852259635925293},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3856751620769501},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3709510564804077},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.3694359362125397},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1675669252872467},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08066973090171814},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"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/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.21437/iberspeech.2022-27","is_oa":false,"landing_page_url":"https://doi.org/10.21437/iberspeech.2022-27","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IberSPEECH 2022","raw_type":"proceedings-article"},{"id":"pmh:oai:pure.atira.dk:openaire/9578dde1-bec6-40b5-8efd-ee47458bb2db","is_oa":true,"landing_page_url":"https://vbn.aau.dk/da/publications/9578dde1-bec6-40b5-8efd-ee47458bb2db","pdf_url":null,"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":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Espejo, I L, Tan, Z-H & Jensen, J 2022, An Experimental Study on Light Speech Features for Small-Footprint Keyword Spotting. in IberSPEECH 2022. IberSPEECH 2022, Granada, Spain, 14/11/2022. https://doi.org/10.21437/IberSPEECH.2022-27","raw_type":"info:eu-repo/semantics/publishedVersion"}],"best_oa_location":{"id":"pmh:oai:pure.atira.dk:openaire/9578dde1-bec6-40b5-8efd-ee47458bb2db","is_oa":true,"landing_page_url":"https://vbn.aau.dk/da/publications/9578dde1-bec6-40b5-8efd-ee47458bb2db","pdf_url":null,"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":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Espejo, I L, Tan, Z-H & Jensen, J 2022, An Experimental Study on Light Speech Features for Small-Footprint Keyword Spotting. in IberSPEECH 2022. IberSPEECH 2022, Granada, Spain, 14/11/2022. https://doi.org/10.21437/IberSPEECH.2022-27","raw_type":"info:eu-repo/semantics/publishedVersion"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W2139983529","https://openalex.org/W2148154194","https://openalex.org/W2191779130","https://openalex.org/W2194775991","https://openalex.org/W2483244395","https://openalex.org/W2797583228","https://openalex.org/W2928062709","https://openalex.org/W2964187693","https://openalex.org/W2964250984","https://openalex.org/W2972792496","https://openalex.org/W2972869267","https://openalex.org/W3015118013","https://openalex.org/W3015273269","https://openalex.org/W3019293601","https://openalex.org/W3114848746","https://openalex.org/W3134145319","https://openalex.org/W3176579201","https://openalex.org/W3198796966","https://openalex.org/W4206567542","https://openalex.org/W4287026017","https://openalex.org/W4287113025","https://openalex.org/W4287123098","https://openalex.org/W4297775537","https://openalex.org/W4309147820"],"related_works":["https://openalex.org/W4391266561","https://openalex.org/W1972612950","https://openalex.org/W4382052966","https://openalex.org/W4205299633","https://openalex.org/W2034940213","https://openalex.org/W3120526039","https://openalex.org/W4287394695","https://openalex.org/W4317383455","https://openalex.org/W2548511587","https://openalex.org/W4246774824"],"abstract_inverted_index":{"Keyword":[0],"spotting":[1],"(KWS)":[2],"is,":[3],"in":[4,49,56,82,120],"many":[5],"instances,":[6],"intended":[7],"to":[8,35,146],"run":[9],"on":[10,95,108],"smart":[11],"electronic":[12],"devices":[13],"characterized":[14],"by":[15,129],"limited":[16],"computational":[17,21],"resources.":[18],"To":[19],"meet":[20],"constraints,":[22],"a":[23,61,92,102,130,147],"series":[24],"of":[25,38,41,65,68,72,98,101,118,132,150],"techniques":[26],"\u2014ranging":[27],"from":[28],"feature":[29,78,88,124],"and":[30,43,74,156],"acoustic":[31,105],"model":[32,106],"parameter":[33],"quantization":[34],"the":[36,39,50,66,69,87,96,116],"reduction":[37,67,144],"number":[40,97,117,149],"model<br/>parameters":[42],"required":[44],"multiplications\u2014":[45],"has":[46,91],"been":[47],"explored":[48],"literature.":[51],"With":[52],"this":[53,57,142],"same":[54],"aim,":[55],"paper,":[58],"we":[59],"study":[60],"straightforward":[62],"alternative":[63],"consisting":[64],"spectro/cepstro-temporal":[70],"resolution":[71],"log-Mel":[73],"Melfrequency":[75],"cepstral":[76],"coefficient":[77],"matrices":[79,125],"commonly":[80,121],"employed":[81],"KWS.":[83],"We":[84],"show":[85],"that":[86,115],"matrix":[89],"size":[90,143],"strong":[93],"impact":[94],"multiplications/energy":[99,151],"consumption":[100],"state-of-the-art":[103],"KWS":[104,137],"based":[107],"convolutional":[109],"neural":[110],"network.":[111],"Experimental":[112],"results":[113],"demonstrate":[114],"elements":[119],"used":[122],"speech":[123],"can":[126],"be":[127],"reduced":[128],"factor":[131],"8":[133],"while":[134],"essentially":[135],"maintaining":[136],"performance.":[138],"Even":[139],"more":[140],"interestingly,":[141],"leads":[145],"9.6\u00d7":[148],"consumption,":[152],"4.0\u00d7":[153],"training":[154],"time":[155,159],"3.7\u00d7":[157],"inference":[158],"reduction.":[160]},"counts_by_year":[{"year":2023,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}