{"id":{"repo_id":"rowan","oai_identifier":"oai:rdw.rowan.edu:etd-3046"},"canonical_url":"https://search.ndltd.org/etd/rowan/oai:rdw.rowan.edu:etd-3046","repository":{"repo_id":"rowan","name":"Rowan University","base_url":"https://rdw.rowan.edu/do/oai/"},"display":{"title":"Robust speaker recognition in the presence of speech coding distortion","abstract":"For wireless remote access security, forensics, border control and surveillance applications, there is an emerging need for biometric speaker recognition systems to be robust to speech coding distortion. This thesis examines the robustness issue for three coders, namely, the ITU-T 6.3 kilobits per second (kbps) G.723.1, the ITU-T 8 kbps G.729 and the 12.2 kbps 3GPP GSM-AMR coder. Both speaker identiﬁcation (SI) and speaker veriﬁcation (SV) systems are considered and use a Gaussian mixture model (GMM) classiﬁer. The systems are trained on clean speech and tested on the decoded speech. To mitigate the performance loss due to mismatched training and testing conditions, four robust features, two enhancement approaches and feature (SI) and score (SV) based fusion strategies are implemented. The ﬁrst proposed novel enhancement method is feature compensation based on the afﬁne transform and is used to map the features from the test scenario to the train scenario. The second is the McCree signal enhancement approach based on the spectral envelope information. A detailed two-way analysis of variance (ANOVA) supplemented with a multiple comparison test is performed in order to show statistical significance in application of these enhancement methods.","abstract_html":"For wireless remote access security, forensics, border control and surveillance applications, there is an emerging need for biometric speaker recognition systems to be robust to speech coding distortion. This thesis examines the robustness issue for three coders, namely, the ITU-T 6.3 kilobits per second (kbps) G.723.1, the ITU-T 8 kbps G.729 and the 12.2 kbps 3GPP GSM-AMR coder. Both speaker identiﬁcation (SI) and speaker veriﬁcation (SV) systems are considered and use a Gaussian mixture model (GMM) classiﬁer. The systems are trained on clean speech and tested on the decoded speech. To mitigate the performance loss due to mismatched training and testing conditions, four robust features, two enhancement approaches and feature (SI) and score (SV) based fusion strategies are implemented. The ﬁrst proposed novel enhancement method is feature compensation based on the afﬁne transform and is used to map the features from the test scenario to the train scenario. The second is the McCree signal enhancement approach based on the spectral envelope information. A detailed two-way analysis of variance (ANOVA) supplemented with a multiple comparison test is performed in order to show statistical significance in application of these enhancement methods.","abstract_has_math":false,"creators":["Mudrosky, Robert Walter"],"institution":null,"degree_name":"M.S. Electrical and Computer Engineering","degree_level":"Thesis","degree_discipline":"Electrical and Computer Engineering","degree_department":null,"school":null,"contributors":["Ramachandran, Ravi","Thayasivam, Umashanger","Schmalzel, John"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:58:20Z","subjects":["affine transform","feature enhancement","GMM classifier","speaker recognition","speech coding distortion","Automatic speech recognition","Speech processing systems","Electrical and Computer Engineering"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://rdw.rowan.edu/etd/2046","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://rdw.rowan.edu/etd/2046","record_label":"Repository record","record_kind":"record","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://rdw.rowan.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Ardw.rowan.edu%3Aetd-3046","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Ramachandran, Ravi","Thayasivam, Umashanger","Schmalzel, John"]},{"key":"dc:creator","label":"Author","values":["Mudrosky, Robert Walter"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2016-08-24T07:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2016-08-23T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical and Computer Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S. Electrical and Computer Engineering"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["affine transform","feature enhancement","GMM classifier","speaker recognition","speech coding distortion","Automatic speech recognition","Speech processing systems","Electrical and Computer Engineering"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://rdw.rowan.edu/etd/2046"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["For wireless remote access security, forensics, border control and surveillance applications, there is an emerging need for biometric speaker recognition systems to be robust to speech coding distortion. This thesis examines the robustness issue for three coders, namely, the ITU-T 6.3 kilobits per second (kbps) G.723.1, the ITU-T 8 kbps G.729 and the 12.2 kbps 3GPP GSM-AMR coder. Both speaker identiﬁcation (SI) and speaker veriﬁcation (SV) systems are considered and use a Gaussian mixture model (GMM) classiﬁer. The systems are trained on clean speech and tested on the decoded speech. To mitigate the performance loss due to mismatched training and testing conditions, four robust features, two enhancement approaches and feature (SI) and score (SV) based fusion strategies are implemented. The ﬁrst proposed novel enhancement method is feature compensation based on the afﬁne transform and is used to map the features from the test scenario to the train scenario. The second is the McCree signal enhancement approach based on the spectral envelope information. A detailed two-way analysis of variance (ANOVA) supplemented with a multiple comparison test is performed in order to show statistical significance in application of these enhancement methods."]},{"key":"dc:title","label":"Title","values":["Robust speaker recognition in the presence of speech coding distortion"]}]}],"canonical_facts":{"dc:contributor":["Ramachandran, Ravi","Thayasivam, Umashanger","Schmalzel, John"],"dc:creator":["Mudrosky, Robert Walter"],"dc:date.available":["2016-08-24T07:00:00Z"],"dc:date.created":["2016-08-23T07:00:00Z"],"dc:description.abstract":["For wireless remote access security, forensics, border control and surveillance applications, there is an emerging need for biometric speaker recognition systems to be robust to speech coding distortion. This thesis examines the robustness issue for three coders, namely, the ITU-T 6.3 kilobits per second (kbps) G.723.1, the ITU-T 8 kbps G.729 and the 12.2 kbps 3GPP GSM-AMR coder. Both speaker identiﬁcation (SI) and speaker veriﬁcation (SV) systems are considered and use a Gaussian mixture model (GMM) classiﬁer. The systems are trained on clean speech and tested on the decoded speech. To mitigate the performance loss due to mismatched training and testing conditions, four robust features, two enhancement approaches and feature (SI) and score (SV) based fusion strategies are implemented. The ﬁrst proposed novel enhancement method is feature compensation based on the afﬁne transform and is used to map the features from the test scenario to the train scenario. The second is the McCree signal enhancement approach based on the spectral envelope information. A detailed two-way analysis of variance (ANOVA) supplemented with a multiple comparison test is performed in order to show statistical significance in application of these enhancement methods."],"dc:identifier":["https://rdw.rowan.edu/etd/2046"],"dc:subject":["affine transform","feature enhancement","GMM classifier","speaker recognition","speech coding distortion","Automatic speech recognition","Speech processing systems","Electrical and Computer Engineering"],"dc:title":["Robust speaker recognition in the presence of speech coding distortion"],"thesis:degree_discipline":["Electrical and Computer Engineering"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S. Electrical and Computer Engineering"]},"updated_at":"2026-11-19T08:58:20Z"}