{"id":"https://openalex.org/W3006710048","doi":"https://doi.org/10.1109/asru46091.2019.9003738","title":"Orthogonality Constrained Multi-Head Attention for Keyword Spotting","display_name":"Orthogonality Constrained Multi-Head Attention for Keyword Spotting","publication_year":2019,"publication_date":"2019-12-01","ids":{"openalex":"https://openalex.org/W3006710048","doi":"https://doi.org/10.1109/asru46091.2019.9003738","mag":"3006710048"},"language":"en","primary_location":{"id":"doi:10.1109/asru46091.2019.9003738","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru46091.2019.9003738","pdf_url":null,"source":{"id":"https://openalex.org/S4306498489","display_name":"2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5017854079","display_name":"Mingu Lee","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mingu Lee","raw_affiliation_strings":["Qualcomm AI Research, Qualcomm Korea, YH"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Qualcomm AI Research, Qualcomm Korea, YH","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101713446","display_name":"Jinkyu Lee","orcid":"https://orcid.org/0000-0002-9838-4908"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinkyu Lee","raw_affiliation_strings":["Qualcomm AI Research, Qualcomm Korea, YH"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Qualcomm AI Research, Qualcomm Korea, YH","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047478012","display_name":"Hye Jin Jang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hye Jin Jang","raw_affiliation_strings":["Qualcomm AI Research, Qualcomm Korea, YH"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Qualcomm AI Research, Qualcomm Korea, YH","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081897616","display_name":"Byeonggeun Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Byeonggeun Kim","raw_affiliation_strings":["Qualcomm AI Research, Qualcomm Korea, YH"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Qualcomm AI Research, Qualcomm Korea, YH","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038745666","display_name":"Wonil Chang","orcid":"https://orcid.org/0000-0002-4457-4917"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wonil Chang","raw_affiliation_strings":["Qualcomm AI Research, Qualcomm Korea, YH"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Qualcomm AI Research, Qualcomm Korea, YH","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089149225","display_name":"Kyuwoong Hwang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kyuwoong Hwang","raw_affiliation_strings":["Qualcomm AI Research, Qualcomm Korea, YH"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Qualcomm AI Research, Qualcomm Korea, YH","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.3015,"has_fulltext":false,"cited_by_count":12,"citation_normalized_percentile":{"value":0.86131169,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"86","last_page":"92"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9994999766349792,"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":0.9994999766349792,"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/T10028","display_name":"Topic Modeling","score":0.9987000226974487,"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/T13083","display_name":"Advanced Text Analysis Techniques","score":0.9973999857902527,"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/keyword-spotting","display_name":"Keyword spotting","score":0.9300973415374756},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7912842035293579},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.6525919437408447},{"id":"https://openalex.org/keywords/spotting","display_name":"Spotting","score":0.6240371465682983},{"id":"https://openalex.org/keywords/orthogonality","display_name":"Orthogonality","score":0.5848819017410278},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5815333127975464},{"id":"https://openalex.org/keywords/sequence-labeling","display_name":"Sequence labeling","score":0.5705212950706482},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.461725652217865},{"id":"https://openalex.org/keywords/head","display_name":"Head (geology)","score":0.44606369733810425},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.4378378391265869},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.4215097427368164},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.42099684476852417},{"id":"https://openalex.org/keywords/feature-vector","display_name":"Feature vector","score":0.4158231019973755},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.37822267413139343},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.10798922181129456}],"concepts":[{"id":"https://openalex.org/C2781213101","wikidata":"https://www.wikidata.org/wiki/Q6398558","display_name":"Keyword spotting","level":2,"score":0.9300973415374756},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7912842035293579},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.6525919437408447},{"id":"https://openalex.org/C2779506182","wikidata":"https://www.wikidata.org/wiki/Q7580141","display_name":"Spotting","level":2,"score":0.6240371465682983},{"id":"https://openalex.org/C17137986","wikidata":"https://www.wikidata.org/wiki/Q215067","display_name":"Orthogonality","level":2,"score":0.5848819017410278},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5815333127975464},{"id":"https://openalex.org/C35639132","wikidata":"https://www.wikidata.org/wiki/Q7452468","display_name":"Sequence labeling","level":3,"score":0.5705212950706482},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.461725652217865},{"id":"https://openalex.org/C2780312720","wikidata":"https://www.wikidata.org/wiki/Q5689100","display_name":"Head (geology)","level":2,"score":0.44606369733810425},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.4378378391265869},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.4215097427368164},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.42099684476852417},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.4158231019973755},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.37822267413139343},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.10798922181129456},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C114793014","wikidata":"https://www.wikidata.org/wiki/Q52109","display_name":"Geomorphology","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},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.0},{"id":"https://openalex.org/C54355233","wikidata":"https://www.wikidata.org/wiki/Q7162","display_name":"Genetics","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C127313418","wikidata":"https://www.wikidata.org/wiki/Q1069","display_name":"Geology","level":0,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asru46091.2019.9003738","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru46091.2019.9003738","pdf_url":null,"source":{"id":"https://openalex.org/S4306498489","display_name":"2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"}],"best_oa_location":null,"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/W2034940213","https://openalex.org/W2056986588","https://openalex.org/W2407023693","https://openalex.org/W2478006605","https://openalex.org/W2558258029","https://openalex.org/W2602634800","https://openalex.org/W2888641632","https://openalex.org/W2892009249","https://openalex.org/W2913718171","https://openalex.org/W2944255943","https://openalex.org/W2953219395","https://openalex.org/W2962707338","https://openalex.org/W2962822108","https://openalex.org/W2962824709","https://openalex.org/W2962959915","https://openalex.org/W2963403868","https://openalex.org/W2963414149","https://openalex.org/W2963881567","https://openalex.org/W2964121744","https://openalex.org/W3103005696","https://openalex.org/W4294621971","https://openalex.org/W4385245566","https://openalex.org/W6739901393","https://openalex.org/W6744661987"],"related_works":["https://openalex.org/W2918559346","https://openalex.org/W3119978414","https://openalex.org/W2114097550","https://openalex.org/W2545741539","https://openalex.org/W2516975559","https://openalex.org/W3206647229","https://openalex.org/W4286904253","https://openalex.org/W2000885660","https://openalex.org/W1969408022","https://openalex.org/W1989658893"],"abstract_inverted_index":{"Multi-head":[0],"attention":[1,14,36,54,62,80,100,134],"mechanism":[2,81],"is":[3],"capable":[4],"of":[5,51],"learning":[6],"various":[7],"representations":[8,106,139],"from":[9,107],"sequential":[10],"data":[11],"while":[12],"paying":[13],"to":[15,103,136],"different":[16,105],"subsequences,":[17,28,109],"e.g.,":[18],"word-pieces":[19],"or":[20],"syllables":[21],"in":[22,82,111],"a":[23,34,48,75],"spoken":[24],"word.":[25],"From":[26],"the":[27,40,52,61,99,155,161,166],"it":[29],"retrieves":[30],"richer":[31],"information":[32,116],"than":[33],"single-head":[35],"which":[37,92,110,143],"only":[38],"summarizes":[39],"whole":[41],"sequence":[42,119],"into":[43],"one":[44],"context":[45],"vector.":[46],"However,":[47],"naive":[49],"use":[50],"multi-head":[53,79],"does":[55],"not":[56],"guarantee":[57],"such":[58,121],"richness":[59],"as":[60,122],"heads":[63,101],"may":[64],"have":[65,137],"positional":[66,94],"and":[67,95],"representational":[68],"redundancy.":[69],"In":[70,126],"this":[71],"paper,":[72],"we":[73],"propose":[74],"regularization":[76,90,131,157],"technique":[77,158],"for":[78,165],"an":[83],"end-to-end":[84],"neural":[85],"keyword":[86,141,162,167],"spotting":[87,163],"system.":[88],"Augmenting":[89],"terms":[91],"penalize":[93],"contextual":[96,129],"non-orthogonality":[97,130],"between":[98],"encourages":[102,132],"output":[104],"separate":[108],"turn":[112],"enables":[113],"leveraging":[114],"structured":[115],"without":[117],"explicit":[118],"models":[120],"hidden":[123],"Markov":[124],"models.":[125],"addition,":[127],"intra-head":[128],"each":[133],"head":[135],"similar":[138],"across":[140],"examples,":[142],"helps":[144],"classification":[145],"by":[146],"reducing":[147],"feature":[148],"variability.":[149],"The":[150],"experimental":[151],"results":[152],"demonstrate":[153],"that":[154],"proposed":[156],"significantly":[159],"improves":[160],"performance":[164],"\u201cHey":[168],"Snapdragon\u201d.":[169]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":4},{"year":2020,"cited_by_count":2}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}