{"created":"2023-05-15T12:34:45.758748+00:00","id":3399,"links":{},"metadata":{"_buckets":{"deposit":"dfc36f65-86c9-4da2-b61a-35284b1be2b4"},"_deposit":{"created_by":3,"id":"3399","owners":[3],"pid":{"revision_id":0,"type":"depid","value":"3399"},"status":"published"},"_oai":{"id":"oai:nitech.repo.nii.ac.jp:00003399","sets":["31"]},"author_link":["464","462"],"item_5_biblio_info_5":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicIssueDates":{"bibliographicIssueDate":"2008","bibliographicIssueDateType":"Issued"},"bibliographicPageEnd":"580","bibliographicPageStart":"577","bibliographic_titles":[{"bibliographic_title":"INTERSPEECH 2008 9th Annual Conference of the International Speech Communication Association"}]}]},"item_5_description_16":{"attribute_name":"フォーマット","attribute_value_mlt":[{"subitem_description":"application/pdf","subitem_description_type":"Other"}]},"item_5_description_4":{"attribute_name":"内容記述","attribute_value_mlt":[{"subitem_description":"A minimum generation error (MGE) criterion had been proposedto solve the issues related to maximum likelihood (ML)based HMM training in HMM-based speech synthesis. In thispaper, we improve the MGE criterion by imposing a log spectraldistortion (LSD) instead of the Euclidean distance to define thegeneration error between the original and generated line spectralpair (LSP) coefficients. Moreover, we investigate the effect ofdifferent sampling strategies to calculate the integration of theLSD function. From the experimental results, using the LSDscalculated by sampling at LSPs achieved the best performance,and the quality of synthesized speech after the MGE-LSD trainingwas improved over the original MGE training.","subitem_description_type":"Other"},{"subitem_description":"Brisbane, AustraliaSeptember 22-26, 2008","subitem_description_type":"Other"}]},"item_5_full_name_3":{"attribute_name":"著者別名","attribute_value_mlt":[{"affiliations":[{"affiliationNames":[{"affiliationName":"","lang":"ja"}],"nameIdentifiers":[{"nameIdentifier":"","nameIdentifierScheme":"ISNI","nameIdentifierURI":"http://www.isni.org/isni/"}]}],"familyNames":[{"familyName":"Tokuda","familyNameLang":"en"},{"familyName":"徳田","familyNameLang":"ja"},{"familyName":"トクダ","familyNameLang":"ja-Kana"}],"givenNames":[{"givenName":"Keiichi","givenNameLang":"en"},{"givenName":"恵一","givenNameLang":"ja"},{"givenName":"ケイイチ","givenNameLang":"ja-Kana"}],"nameIdentifiers":[{"nameIdentifier":"464","nameIdentifierScheme":"WEKO"},{"nameIdentifier":"1000020217483","nameIdentifierScheme":"NRID","nameIdentifierURI":"http://rns.nii.ac.jp/nr/1000020217483"}],"names":[{"name":"Tokuda, Keiichi","nameLang":"en"},{"name":"徳田, 恵一","nameLang":"ja"},{"name":"トクダ, ケイイチ","nameLang":"ja-Kana"}]}]},"item_5_publisher_6":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"International Speech Communication Association"}]},"item_5_relation_12":{"attribute_name":"関連サイト","attribute_value_mlt":[{"subitem_relation_name":[{"subitem_relation_name_text":"http://www.isca-speech.org/archive/interspeech_2008/i08_0577.html"}],"subitem_relation_type_id":{"subitem_relation_type_id_text":"http://www.isca-speech.org/archive/interspeech_2008/i08_0577.html","subitem_relation_type_select":"URI"}}]},"item_5_version_type_13":{"attribute_name":"著者版フラグ","attribute_value_mlt":[{"subitem_version_resource":"http://purl.org/coar/version/c_970fb48d4fbd8a85","subitem_version_type":"VoR"}]},"item_creator":{"attribute_name":"著者","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Wu, Yi-Jian","creatorNameLang":"en"},{"creatorName":"Tokuda, Keiichi","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_files":{"attribute_name":"ファイル情報","attribute_type":"file","attribute_value_mlt":[{"accessrole":"open_date","date":[{"dateType":"Available","dateValue":"2017-01-17"}],"displaytype":"detail","filename":"Interspeech2008_wu.pdf","filesize":[{"value":"247.2 kB"}],"format":"application/pdf","licensetype":"license_note","mimetype":"application/pdf","url":{"label":"本文_fulltext","url":"https://nitech.repo.nii.ac.jp/record/3399/files/Interspeech2008_wu.pdf"},"version_id":"04ab978e-1366-42f8-9e18-4a45c186ee03"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"eng"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourcetype":"conference paper","resourceuri":"http://purl.org/coar/resource_type/c_5794"}]},"item_title":"Minimum generation error training with direct log spectral distortion on LSPs for HMM-based speech synthesis","item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"Minimum generation error training with direct log spectral distortion on LSPs for HMM-based speech synthesis","subitem_title_language":"en"}]},"item_type_id":"5","owner":"3","path":["31"],"pubdate":{"attribute_name":"公開日","attribute_value":"2012-11-07"},"publish_date":"2012-11-07","publish_status":"0","recid":"3399","relation_version_is_last":true,"title":["Minimum generation error training with direct log spectral distortion on LSPs for HMM-based speech synthesis"],"weko_creator_id":"3","weko_shared_id":3},"updated":"2023-07-10T03:01:06.999957+00:00"}