{"id":"https://openalex.org/W4223531842","doi":"https://doi.org/10.21437/interspeech.2022-10055","title":"Scoring of Large-Margin Embeddings for Speaker Verification: Cosine or PLDA?","display_name":"Scoring of Large-Margin Embeddings for Speaker Verification: Cosine or PLDA?","publication_year":2022,"publication_date":"2022-09-16","ids":{"openalex":"https://openalex.org/W4223531842","doi":"https://doi.org/10.21437/interspeech.2022-10055"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2022-10055","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10055","pdf_url":null,"source":{"id":"https://openalex.org/S4363604309","display_name":"Interspeech 2022","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2022","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/A5101991100","display_name":"Qiongqiong Wang","orcid":"https://orcid.org/0000-0002-9903-0618"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Qiongqiong Wang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004287909","display_name":"Kong Aik Lee","orcid":"https://orcid.org/0000-0001-9133-3000"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kong Aik Lee","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5013047791","display_name":"Tianchi Liu","orcid":"https://orcid.org/0000-0003-3472-0703"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tianchi Liu","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5101991100"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.1051,"has_fulltext":false,"cited_by_count":20,"citation_normalized_percentile":{"value":0.88994328,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"600","last_page":"604"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9993000030517578,"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":0.9993000030517578,"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":0.9868000149726868,"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":0.9620000123977661,"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/softmax-function","display_name":"Softmax function","score":0.8000781536102295},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7181135416030884},{"id":"https://openalex.org/keywords/speaker-verification","display_name":"Speaker verification","score":0.6598387360572815},{"id":"https://openalex.org/keywords/margin","display_name":"Margin (machine learning)","score":0.65384840965271},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.5975124835968018},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5891176462173462},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5229190587997437},{"id":"https://openalex.org/keywords/cosine-similarity","display_name":"Cosine similarity","score":0.4942786693572998},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.4940965473651886},{"id":"https://openalex.org/keywords/parametric-statistics","display_name":"Parametric statistics","score":0.466043621301651},{"id":"https://openalex.org/keywords/trigonometric-functions","display_name":"Trigonometric functions","score":0.4348568916320801},{"id":"https://openalex.org/keywords/probabilistic-logic","display_name":"Probabilistic logic","score":0.43293529748916626},{"id":"https://openalex.org/keywords/cross-entropy","display_name":"Cross entropy","score":0.4319096505641937},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.40922796726226807},{"id":"https://openalex.org/keywords/speaker-recognition","display_name":"Speaker recognition","score":0.33340317010879517},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.23739734292030334},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.1653113067150116},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.10464704036712646}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.8000781536102295},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7181135416030884},{"id":"https://openalex.org/C2982762665","wikidata":"https://www.wikidata.org/wiki/Q1145189","display_name":"Speaker verification","level":3,"score":0.6598387360572815},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.65384840965271},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.5975124835968018},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5891176462173462},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5229190587997437},{"id":"https://openalex.org/C2780762811","wikidata":"https://www.wikidata.org/wiki/Q1784941","display_name":"Cosine similarity","level":3,"score":0.4942786693572998},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.4940965473651886},{"id":"https://openalex.org/C117251300","wikidata":"https://www.wikidata.org/wiki/Q1849855","display_name":"Parametric statistics","level":2,"score":0.466043621301651},{"id":"https://openalex.org/C178009071","wikidata":"https://www.wikidata.org/wiki/Q93344","display_name":"Trigonometric functions","level":2,"score":0.4348568916320801},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.43293529748916626},{"id":"https://openalex.org/C167981619","wikidata":"https://www.wikidata.org/wiki/Q1685498","display_name":"Cross entropy","level":3,"score":0.4319096505641937},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.40922796726226807},{"id":"https://openalex.org/C133892786","wikidata":"https://www.wikidata.org/wiki/Q1145189","display_name":"Speaker recognition","level":2,"score":0.33340317010879517},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.23739734292030334},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.1653113067150116},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.10464704036712646},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2022-10055","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10055","pdf_url":null,"source":{"id":"https://openalex.org/S4363604309","display_name":"Interspeech 2022","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2022","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.6299999952316284,"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4367154275","https://openalex.org/W2913125146","https://openalex.org/W66821593","https://openalex.org/W2947728370","https://openalex.org/W1521299571","https://openalex.org/W2911303748","https://openalex.org/W3141593045","https://openalex.org/W204267554","https://openalex.org/W2134501921","https://openalex.org/W4252590334"],"abstract_inverted_index":{"The":[0],"emergence":[1],"of":[2,49,131,140],"large-margin":[3,89,100],"softmax":[4,101],"cross-entropy":[5,51,102],"losses":[6,52,103],"in":[7,81,108,172],"training":[8],"deep":[9],"speaker":[10,29,67],"embedding":[11],"neural":[12],"networks":[13,96],"has":[14],"triggered":[15],"a":[16,23,56,105,138],"gradual":[17],"shift":[18,57],"from":[19],"parametric":[20,32],"back-ends":[21,33,63],"to":[22,54,60],"simpler":[24],"cosine":[25,167],"similarity":[26],"measure":[27],"for":[28,66],"verification.":[30,68],"Popular":[31],"include":[34],"the":[35,47,73,82,93,112,123,129,132,143],"probabilistic":[36],"linear":[37],"discriminant":[38],"analysis":[39],"(PLDA)":[40],"and":[41,58,84,145,157,174,178],"its":[42],"variants.":[43],"This":[44],"paper":[45],"investigates":[46],"properties":[48],"margin-based":[50],"leading":[53],"such":[55],"aims":[59],"find":[61],"scoring":[62,168],"best":[64],"suited":[65],"In":[69,116],"addition,":[70],"we":[71,119],"revisit":[72],"pre-processing":[74],"techniques":[75],"which":[76],"have":[77],"been":[78],"widely":[79],"used":[80],"past":[83],"assess":[85],"their":[86],"effectiveness":[87],"on":[88,92,142],"embeddings.":[90],"Experiments":[91],"state-of-the-art":[94],"ECAPA-TDNN":[95],"trained":[97],"with":[98,150,170],"various":[99],"show":[104],"substantial":[106],"increment":[107],"intra-speaker":[109],"compactness":[110],"making":[111],"conventional":[113],"PLDA":[114],"superfluous.":[115],"this":[117],"regard,":[118],"found":[120],"that":[121],"constraining":[122],"within-speaker":[124],"covariance":[125],"matrix":[126],"could":[127],"improve":[128],"performance":[130],"PLDA.":[133],"It":[134,164],"is":[135],"demonstrated":[136],"through":[137],"series":[139],"experiments":[141],"VoxCeleb-1":[144],"SITW":[146],"core-core":[147],"test":[148],"sets":[149],"40.8%":[151],"equal":[152],"error":[153],"rate":[154],"(EER)":[155],"reduction":[156],"35.1%":[158],"minimum":[159],"detection":[160],"cost":[161],"(minDCF)":[162],"reduction.":[163],"also":[165],"outperforms":[166],"consistently":[169],"reductions":[171],"EER":[173],"minDCF":[175],"by":[176],"10.9%":[177],"4.9%,":[179],"respectively.":[180]},"counts_by_year":[{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":9}],"updated_date":"2026-03-05T09:29:38.588285","created_date":"2025-10-10T00:00:00"}