{"id":"https://openalex.org/W4402673686","doi":"https://doi.org/10.1109/taslp.2024.3463430","title":"On the Quantization of Neural Models for Speaker Verification","display_name":"On the Quantization of Neural Models for Speaker Verification","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W4402673686","doi":"https://doi.org/10.1109/taslp.2024.3463430"},"language":"en","primary_location":{"id":"doi:10.1109/taslp.2024.3463430","is_oa":false,"landing_page_url":"https://doi.org/10.1109/taslp.2024.3463430","pdf_url":null,"source":{"id":"https://openalex.org/S4210169297","display_name":"IEEE/ACM Transactions on Audio Speech and Language Processing","issn_l":"2329-9290","issn":["2329-9290","2329-9304"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE/ACM Transactions on Audio, Speech, and Language Processing","raw_type":"journal-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/A5104342086","display_name":"Vishal Kumar","orcid":"https://orcid.org/0009-0000-1549-6350"},"institutions":[{"id":"https://openalex.org/I110675161","display_name":"Infosys (India)","ror":"https://ror.org/03bs18y54","country_code":"IN","type":"company","lineage":["https://openalex.org/I110675161"]},{"id":"https://openalex.org/I119939252","display_name":"Indraprastha Institute of Information Technology Delhi","ror":"https://ror.org/03vfp4g33","country_code":"IN","type":"education","lineage":["https://openalex.org/I119939252"]}],"countries":["IN"],"is_corresponding":true,"raw_author_name":"Vishal Kumar","raw_affiliation_strings":["Infosys Centre for AI and CSE Department, Indraprastha Institute of Information Technology Delhi, New Delhi, India"],"affiliations":[{"raw_affiliation_string":"Infosys Centre for AI and CSE Department, Indraprastha Institute of Information Technology Delhi, New Delhi, India","institution_ids":["https://openalex.org/I119939252","https://openalex.org/I110675161"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075520691","display_name":"Vinayak Abrol","orcid":"https://orcid.org/0000-0001-8149-8151"},"institutions":[{"id":"https://openalex.org/I119939252","display_name":"Indraprastha Institute of Information Technology Delhi","ror":"https://ror.org/03vfp4g33","country_code":"IN","type":"education","lineage":["https://openalex.org/I119939252"]},{"id":"https://openalex.org/I110675161","display_name":"Infosys (India)","ror":"https://ror.org/03bs18y54","country_code":"IN","type":"company","lineage":["https://openalex.org/I110675161"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Vinayak Abrol","raw_affiliation_strings":["Infosys Centre for AI and CSE Department, Indraprastha Institute of Information Technology Delhi, New Delhi, India"],"affiliations":[{"raw_affiliation_string":"Infosys Centre for AI and CSE Department, Indraprastha Institute of Information Technology Delhi, New Delhi, India","institution_ids":["https://openalex.org/I119939252","https://openalex.org/I110675161"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5043551083","display_name":"Mathew Magimai.-Doss","orcid":"https://orcid.org/0000-0002-8714-1409"},"institutions":[{"id":"https://openalex.org/I7495430","display_name":"Idiap Research Institute","ror":"https://ror.org/05932h694","country_code":"CH","type":"facility","lineage":["https://openalex.org/I7495430"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Mathew Magamai Doss","raw_affiliation_strings":["Idiap Research Institute, Martigny, Switzerland"],"affiliations":[{"raw_affiliation_string":"Idiap Research Institute, Martigny, Switzerland","institution_ids":["https://openalex.org/I7495430"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5104342086"],"corresponding_institution_ids":["https://openalex.org/I110675161","https://openalex.org/I119939252"],"apc_list":null,"apc_paid":null,"fwci":0.3626,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.66918448,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":"32","issue":null,"first_page":"4226","last_page":"4236"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9866999983787537,"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":0.9866999983787537,"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":0.9072999954223633,"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/speaker-verification","display_name":"Speaker verification","score":0.6813198328018188},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5684593319892883},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5661719441413879},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.5418283939361572},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3370281457901001},{"id":"https://openalex.org/keywords/speaker-recognition","display_name":"Speaker recognition","score":0.329559862613678},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.10759589076042175}],"concepts":[{"id":"https://openalex.org/C2982762665","wikidata":"https://www.wikidata.org/wiki/Q1145189","display_name":"Speaker verification","level":3,"score":0.6813198328018188},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5684593319892883},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5661719441413879},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.5418283939361572},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3370281457901001},{"id":"https://openalex.org/C133892786","wikidata":"https://www.wikidata.org/wiki/Q1145189","display_name":"Speaker recognition","level":2,"score":0.329559862613678},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.10759589076042175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/taslp.2024.3463430","is_oa":false,"landing_page_url":"https://doi.org/10.1109/taslp.2024.3463430","pdf_url":null,"source":{"id":"https://openalex.org/S4210169297","display_name":"IEEE/ACM Transactions on Audio Speech and Language Processing","issn_l":"2329-9290","issn":["2329-9290","2329-9304"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE/ACM Transactions on Audio, Speech, and Language Processing","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322753","display_name":"Infosys Foundation","ror":"https://ror.org/03bs18y54"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":44,"referenced_works":["https://openalex.org/W2194775991","https://openalex.org/W2242818861","https://openalex.org/W2726515241","https://openalex.org/W2903711666","https://openalex.org/W2928062709","https://openalex.org/W2936774411","https://openalex.org/W2938358845","https://openalex.org/W2963480671","https://openalex.org/W2967733054","https://openalex.org/W3024869864","https://openalex.org/W3034368386","https://openalex.org/W3035183452","https://openalex.org/W3035332806","https://openalex.org/W3035452548","https://openalex.org/W3040522390","https://openalex.org/W3128299679","https://openalex.org/W3138154797","https://openalex.org/W3205878676","https://openalex.org/W4214620071","https://openalex.org/W4225674648","https://openalex.org/W4226426325","https://openalex.org/W4297841773","https://openalex.org/W4304480884","https://openalex.org/W4313559399","https://openalex.org/W4362653163","https://openalex.org/W4385819842","https://openalex.org/W4385823155","https://openalex.org/W4386083031","https://openalex.org/W4394625578","https://openalex.org/W6677580257","https://openalex.org/W6690026940","https://openalex.org/W6693397755","https://openalex.org/W6743969146","https://openalex.org/W6745722055","https://openalex.org/W6748163181","https://openalex.org/W6752784787","https://openalex.org/W6758046424","https://openalex.org/W6760069825","https://openalex.org/W6767597771","https://openalex.org/W6770425567","https://openalex.org/W6791603202","https://openalex.org/W6838398888","https://openalex.org/W6838537176","https://openalex.org/W6842795023"],"related_works":["https://openalex.org/W66821593","https://openalex.org/W4297807400","https://openalex.org/W1491159402","https://openalex.org/W4313854686","https://openalex.org/W321304764","https://openalex.org/W2249138175","https://openalex.org/W1521299571","https://openalex.org/W3162054169","https://openalex.org/W1813780412","https://openalex.org/W2140022733"],"abstract_inverted_index":{"This":[0],"paper":[1],"addresses":[2],"the":[3,60,86,91,105],"sub-optimality":[4],"of":[5,93,107],"current":[6],"post-training":[7],"quantization":[8],"(PTQ)":[9],"and":[10,29,51,65,97,117],"quantization-aware":[11],"training":[12],"(QAT)":[13],"methods":[14],"for":[15,99],"state-of-the-art":[16],"speaker":[17],"verification":[18],"(SV)":[19],"models":[20],"featuring":[21],"intricate":[22],"architectural":[23],"elements":[24],"such":[25],"as":[26],"channel":[27],"aggregation":[28],"squeeze":[30],"excitation":[31],"modules.":[32],"To":[33],"address":[34],"these":[35],"limitations,":[36],"we":[37],"propose":[38],"1)":[39],"a":[40,53],"data-independent":[41],"PTQ":[42],"technique":[43],"employing":[44],"iterative":[45],"low-precision":[46],"calibration":[47],"on":[48,85,125],"pre-trained":[49],"models;":[50],"2)":[52],"data-dependent":[54],"QAT":[55,69],"method":[56,109],"designed":[57],"to":[58],"reduce":[59],"performance":[61,124],"gap":[62],"between":[63],"full-precision":[64,128],"integer":[66],"models.":[67,129],"Our":[68],"involves":[70],"two":[71],"progressive":[72],"stages":[73],"where":[74],"FP-32":[75],"weights":[76],"are":[77],"initially":[78],"transformed":[79],"into":[80],"FP-8,":[81],"adapting":[82],"precision":[83],"based":[84],"gradient":[87],"norm,":[88],"followed":[89],"by":[90],"learning":[92],"quantizer":[94],"parameters":[95],"(scale":[96],"zero-point)":[98],"INT8":[100,118],"conversion.":[101],"Experimental":[102],"validation":[103],"underscores":[104],"ingenuity":[106],"our":[108],"in":[110],"model":[111],"quantization,":[112],"demonstrating":[113],"reduced":[114],"floating-point":[115],"operations":[116],"inference":[119],"time,":[120],"all":[121],"while":[122],"maintaining":[123],"par":[126],"with":[127]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2025-12-27T23:08:20.325037","created_date":"2025-10-10T00:00:00"}