{"id":{"repo_id":"msu","oai_identifier":"oai:d.lib.msu.edu:etd_51270"},"canonical_url":"https://search.ndltd.org/etd/msu/oai:d.lib.msu.edu:etd_51270","repository":{"repo_id":"msu","name":"Michigan State University","base_url":"https://d.lib.msu.edu/oai"},"display":{"title":"Vocal style factorization for effective speaker recognition in affective scenarios","abstract":"The accuracy of automated speaker recognition is negatively impacted by change in emotions in a person's speech. In this thesis, we hypothesize that speaker identity is composed of various vocal style factors that may be learned from unlabeled speech data and re-combined using a neural network architecture to generate holistic speaker identity representations for affective scenarios. In this regard we propose the E-Vector neural network architecture, composed of a 1-D CNN for learning speaker identity features and a vocal style factorization technique for determining vocal styles. Experiments conducted on the MSP-Podcast dataset demonstrate that the proposed architecture improves state-of-the-art speaker recognition accuracy in the affective domain over baseline ECAPA-TDNN speaker recognition models. For instance, the true match rate at a false match rate of 1% improves from 27.6% to 46.2%. Additionally, we provide an analysis between speaker recognition match scores and emotions to identify challenging affective scenarios.","abstract_html":"The accuracy of automated speaker recognition is negatively impacted by change in emotions in a person&#x27;s speech. In this thesis, we hypothesize that speaker identity is composed of various vocal style factors that may be learned from unlabeled speech data and re-combined using a neural network architecture to generate holistic speaker identity representations for affective scenarios. In this regard we propose the E-Vector neural network architecture, composed of a 1-D CNN for learning speaker identity features and a vocal style factorization technique for determining vocal styles. Experiments conducted on the MSP-Podcast dataset demonstrate that the proposed architecture improves state-of-the-art speaker recognition accuracy in the affective domain over baseline ECAPA-TDNN speaker recognition models. For instance, the true match rate at a false match rate of 1% improves from 27.6% to 46.2%. Additionally, we provide an analysis between speaker recognition match scores and emotions to identify challenging affective scenarios.","abstract_has_math":false,"creators":["Sandler, Morgan Lee"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Ross, Arun A.","Kordjamshidi, Parisa","Yan, Qiben"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2023,"date_issued":"2023","date_published":"2023","updated_at":"2026-11-18T18:57:59Z","subjects":["Artificial intelligence","Computer science"],"languages":["English"],"rights":["Attribution 4.0 International"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:51270","isbn:9798379615116","oclc:on1396767666","oclc:1396767666","umi:30494340","local:Sandler_grad.msu_0128N_19864"],"render_values":[{"text":"etd:51270","href":null,"code":true},{"text":"isbn:9798379615116","href":null,"code":true},{"text":"oclc:on1396767666","href":null,"code":true},{"text":"oclc:1396767666","href":null,"code":true},{"text":"umi:30494340","href":null,"code":true},{"text":"local:Sandler_grad.msu_0128N_19864","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/doi:10.25335/nrsx-g437","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/doi:10.25335/nrsx-g437","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://d.lib.msu.edu/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Ad.lib.msu.edu%3Aetd_51270","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Ross, Arun A.","Kordjamshidi, Parisa","Yan, Qiben"]},{"key":"dc:creator","label":"Author","values":["Sandler, Morgan Lee"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2023"]},{"key":"dc:relation","label":"Dc Relation","values":["Electronic Theses & Dissertations"]},{"key":"dc:type","label":"Dc Type","values":["Text","Theses"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Artificial intelligence","Computer science"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["English"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution 4.0 International"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:51270","isbn:9798379615116","oclc:on1396767666","oclc:1396767666","umi:30494340","local:Sandler_grad.msu_0128N_19864","https://doi.org/doi:10.25335/nrsx-g437"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["The accuracy of automated speaker recognition is negatively impacted by change in emotions in a person's speech. In this thesis, we hypothesize that speaker identity is composed of various vocal style factors that may be learned from unlabeled speech data and re-combined using a neural network architecture to generate holistic speaker identity representations for affective scenarios. In this regard we propose the E-Vector neural network architecture, composed of a 1-D CNN for learning speaker identity features and a vocal style factorization technique for determining vocal styles. Experiments conducted on the MSP-Podcast dataset demonstrate that the proposed architecture improves state-of-the-art speaker recognition accuracy in the affective domain over baseline ECAPA-TDNN speaker recognition models. For instance, the true match rate at a false match rate of 1% improves from 27.6% to 46.2%. Additionally, we provide an analysis between speaker recognition match scores and emotions to identify challenging affective scenarios.","Thesis (M.S.)--Michigan State University. Computer Science, 2023","Includes bibliographical references"]},{"key":"dc:format","label":"Dc Format","values":["55 pages","application/pdf"]},{"key":"dc:title","label":"Title","values":["Vocal style factorization for effective speaker recognition in affective scenarios"]}]}],"canonical_facts":{"dc:contributor":["Ross, Arun A.","Kordjamshidi, Parisa","Yan, Qiben"],"dc:creator":["Sandler, Morgan Lee"],"dc:date":["2023"],"dc:description":["The accuracy of automated speaker recognition is negatively impacted by change in emotions in a person's speech. In this thesis, we hypothesize that speaker identity is composed of various vocal style factors that may be learned from unlabeled speech data and re-combined using a neural network architecture to generate holistic speaker identity representations for affective scenarios. In this regard we propose the E-Vector neural network architecture, composed of a 1-D CNN for learning speaker identity features and a vocal style factorization technique for determining vocal styles. Experiments conducted on the MSP-Podcast dataset demonstrate that the proposed architecture improves state-of-the-art speaker recognition accuracy in the affective domain over baseline ECAPA-TDNN speaker recognition models. For instance, the true match rate at a false match rate of 1% improves from 27.6% to 46.2%. Additionally, we provide an analysis between speaker recognition match scores and emotions to identify challenging affective scenarios.","Thesis (M.S.)--Michigan State University. Computer Science, 2023","Includes bibliographical references"],"dc:format":["55 pages","application/pdf"],"dc:identifier":["etd:51270","isbn:9798379615116","oclc:on1396767666","oclc:1396767666","umi:30494340","local:Sandler_grad.msu_0128N_19864","https://doi.org/doi:10.25335/nrsx-g437"],"dc:language":["English"],"dc:relation":["Electronic Theses & Dissertations"],"dc:rights":["Attribution 4.0 International"],"dc:subject":["Artificial intelligence","Computer science"],"dc:title":["Vocal style factorization for effective speaker recognition in affective scenarios"],"dc:type":["Text","Theses"]},"updated_at":"2026-11-18T18:57:59Z"}