{"id":{"repo_id":"gatech","oai_identifier":"oai:repository.gatech.edu:1853/58653"},"canonical_url":"https://search.ndltd.org/etd/gatech/oai:repository.gatech.edu:1853/58653","repository":{"repo_id":"gatech","name":"Georgia Tech","base_url":"https://repository.gatech.edu/server/oai/request"},"display":{"title":"Bayesian adaptation and combination of deep models for automatic speech recognition","abstract":"The objective of the proposed research is to deploy a Bayesian adaptation and combination framework for deep model based automatic speech recognition systems to combat the degradation of the recognition accuracy, which is typically observed under potential mismatched conditions between training and testing. This dissertation addresses the problem in three directions. The first direction is to perform Bayesian adaptation directly on the discriminative deep neural network models. Maximum a posteriori estimation and multi-task learning techniques are employed in the manner of regularization in the deep neural network updating formula. In the second direction, deep neural network is cast into a generative model to better leverage Bayesian techniques. Classic structured maximum a posteriori adaption is adopted by using bottleneck features derived from deep neural networks. In the third direction, a hierarchical Bayesian system combination technique is employed to further enhance the adaptation performance by leveraging the complementarity of the discriminative and generative adaptive models.","abstract_html":"The objective of the proposed research is to deploy a Bayesian adaptation and combination framework for deep model based automatic speech recognition systems to combat the degradation of the recognition accuracy, which is typically observed under potential mismatched conditions between training and testing. This dissertation addresses the problem in three directions. The first direction is to perform Bayesian adaptation directly on the discriminative deep neural network models. Maximum a posteriori estimation and multi-task learning techniques are employed in the manner of regularization in the deep neural network updating formula. In the second direction, deep neural network is cast into a generative model to better leverage Bayesian techniques. Classic structured maximum a posteriori adaption is adopted by using bottleneck features derived from deep neural networks. In the third direction, a hierarchical Bayesian system combination technique is employed to further enhance the adaptation performance by leveraging the complementarity of the discriminative and generative adaptive models.","abstract_has_math":false,"creators":["Huang, Zhen"],"institution":"Georgia Institute of Technology","degree_name":null,"degree_level":"Doctoral","degree_discipline":null,"degree_department":"Electrical and Computer Engineering","school":null,"contributors":[],"advisors":["Lee, Chin-Hui"],"committee_chairs":[],"committee_members":["Juang, Biing-Hwang (Fred)","Clements, Mark A.","Li, Geoffrey Ye","Siniscalchi, Sabato Marco"],"year":2017,"date_issued":"2017-05-25","date_published":"2017-05-25","updated_at":"2026-11-19T05:52:32Z","subjects":["Deep neural networks"],"languages":["en_US"],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/1853/58653","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"http://hdl.handle.net/1853/58653","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://repository.gatech.edu/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Arepository.gatech.edu%3A1853%2F58653","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Lee, Chin-Hui"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Juang, Biing-Hwang (Fred)","Clements, Mark A.","Li, Geoffrey Ye","Siniscalchi, Sabato Marco"]},{"key":"dc:contributor.department","label":"Department","values":["Electrical and Computer Engineering"]},{"key":"dc:contributor.other","label":"Dc Contributor Other","values":["School of Electrical and Computer Engineering","College of Engineering"]},{"key":"dc:creator","label":"Author","values":["Huang, Zhen"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2017-08-17T18:58:55Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2017-08-17T18:58:55Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2017-08"]},{"key":"dc:date.issued","label":"Date","values":["2017-05-25"]},{"key":"dc:date.submitted","label":"Dc Date Submitted","values":["August 2017"]},{"key":"dc:date.updated","label":"Dc Date Updated","values":["2017-08-17T18:58:55Z"]},{"key":"dc:publisher","label":"Institution","values":["Georgia Institute of Technology"]},{"key":"dc:relation.advisorentity","label":"Dc Relation Advisorentity","values":["b35c0c49-bee2-49ad-9d6a-867b4ba8908b"]},{"key":"dc:relation.ispartofseries","label":"Dc Relation Ispartofseries","values":["Doctor of Philosophy with a Major in Electrical and Computer Engineering"]},{"key":"dc:relation.orgunit","label":"Dc Relation Orgunit","values":["5b7adef2-447c-4270-b9fc-846bd76f80f2","7c022d60-21d5-497c-b552-95e489a06569"]},{"key":"dc:relation.seriesentity","label":"Dc Relation Seriesentity","values":["3890cd56-8ecf-424b-8074-8f5ebd8dbe15"]},{"key":"dc:type","label":"Dc Type","values":["Text","Dissertation"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Doctoral"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Deep neural networks"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://hdl.handle.net/1853/58653"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The objective of the proposed research is to deploy a Bayesian adaptation and combination framework for deep model based automatic speech recognition systems to combat the degradation of the recognition accuracy, which is typically observed under potential mismatched conditions between training and testing. This dissertation addresses the problem in three directions. The first direction is to perform Bayesian adaptation directly on the discriminative deep neural network models. Maximum a posteriori estimation and multi-task learning techniques are employed in the manner of regularization in the deep neural network updating formula. In the second direction, deep neural network is cast into a generative model to better leverage Bayesian techniques. Classic structured maximum a posteriori adaption is adopted by using bottleneck features derived from deep neural networks. In the third direction, a hierarchical Bayesian system combination technique is employed to further enhance the adaptation performance by leveraging the complementarity of the discriminative and generative adaptive models."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Ph.D."]},{"key":"dc:format.mimetype","label":"Dc Format Mimetype","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Bayesian adaptation and combination of deep models for automatic speech recognition"]},{"key":"dspace:entity.type","label":"Dspace Entity Type","values":["Publication"]}]}],"canonical_facts":{"dc:contributor.advisor":["Lee, Chin-Hui"],"dc:contributor.committeemember":["Juang, Biing-Hwang (Fred)","Clements, Mark A.","Li, Geoffrey Ye","Siniscalchi, Sabato Marco"],"dc:contributor.department":["Electrical and Computer Engineering"],"dc:contributor.other":["School of Electrical and Computer Engineering","College of Engineering"],"dc:creator":["Huang, Zhen"],"dc:date.accessioned":["2017-08-17T18:58:55Z"],"dc:date.available":["2017-08-17T18:58:55Z"],"dc:date.created":["2017-08"],"dc:date.issued":["2017-05-25"],"dc:date.submitted":["August 2017"],"dc:date.updated":["2017-08-17T18:58:55Z"],"dc:description.abstract":["The objective of the proposed research is to deploy a Bayesian adaptation and combination framework for deep model based automatic speech recognition systems to combat the degradation of the recognition accuracy, which is typically observed under potential mismatched conditions between training and testing. This dissertation addresses the problem in three directions. The first direction is to perform Bayesian adaptation directly on the discriminative deep neural network models. Maximum a posteriori estimation and multi-task learning techniques are employed in the manner of regularization in the deep neural network updating formula. In the second direction, deep neural network is cast into a generative model to better leverage Bayesian techniques. Classic structured maximum a posteriori adaption is adopted by using bottleneck features derived from deep neural networks. In the third direction, a hierarchical Bayesian system combination technique is employed to further enhance the adaptation performance by leveraging the complementarity of the discriminative and generative adaptive models."],"dc:description.degree":["Ph.D."],"dc:format.mimetype":["application/pdf"],"dc:identifier.uri":["http://hdl.handle.net/1853/58653"],"dc:language.iso":["en_US"],"dc:publisher":["Georgia Institute of Technology"],"dc:relation.advisorentity":["b35c0c49-bee2-49ad-9d6a-867b4ba8908b"],"dc:relation.ispartofseries":["Doctor of Philosophy with a Major in Electrical and Computer Engineering"],"dc:relation.orgunit":["5b7adef2-447c-4270-b9fc-846bd76f80f2","7c022d60-21d5-497c-b552-95e489a06569"],"dc:relation.seriesentity":["3890cd56-8ecf-424b-8074-8f5ebd8dbe15"],"dc:subject":["Deep neural networks"],"dc:title":["Bayesian adaptation and combination of deep models for automatic speech recognition"],"dc:type":["Text","Dissertation"],"dspace:entity.type":["Publication"],"thesis:degree_level":["Doctoral"]},"updated_at":"2026-11-19T05:52:32Z"}