{"updated":"2025-01-20T05:29:59.573117+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00177376","sets":["1164:5159:9063:9064"]},"path":["9064"],"owner":"11","recid":"177376","title":["Investigation on an autoregressive recurrent mixture density network for parametric speech synthesis"],"pubdate":{"attribute_name":"公開日","attribute_value":"2017-02-10"},"_buckets":{"deposit":"24219271-92e3-4525-b1a1-b0ae82ce3670"},"_deposit":{"id":"177376","pid":{"type":"depid","value":"177376","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"Investigation on an autoregressive recurrent mixture density network for parametric speech synthesis","author_link":["376219","376218","376220","376217","376216","376221"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"Investigation on an autoregressive recurrent mixture density network for parametric speech synthesis"},{"subitem_title":"Investigation on an autoregressive recurrent mixture density network for parametric speech synthesis","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"音声合成・応用","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2017-02-10","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"National Institute of Informatics"},{"subitem_text_value":"National Institute of Informatics"},{"subitem_text_value":"National Institute of Informatics"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"National Institute of Informatics","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Informatics","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Informatics","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"eng"}]},"item_publisher":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会","subitem_publisher_language":"ja"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/177376/files/IPSJ-SLP17115002.pdf","label":"IPSJ-SLP17115002.pdf"},"date":[{"dateType":"Available","dateValue":"2019-02-10"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP17115002.pdf","filesize":[{"value":"974.8 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"22"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"38ee0ce3-b11f-4627-84c0-2076a508e442","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2017 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Xin, Wang"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Shinji, Takaki"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Junichi, Yamagishi"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Xin, Wang","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Shinji, Takaki","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Junichi, Yamagishi","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10442647","subitem_source_identifier_type":"NCID"}]},"item_4_textarea_12":{"attribute_name":"Notice","attribute_value_mlt":[{"subitem_textarea_value":"SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc."}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_18gh","resourcetype":"technical report"}]},"item_4_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8663","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"Neural-network-based mixture density networks are important tools for acoustic modeling in statistical parametric speech synthesis. Recently we found that incorporating an autoregressive model in a recurrent mixture density network, which is referred to as AR-RMDN, enabled the network to generate quite smooth acoustic data trajectories without using the delta and delta-delta coefficients. More interestingly, the new model generated trajectories with a dynamic range similar to that of the natural data, thus alleviating over-smoothing effect. In this work, after explaining the AR-RMDN from the perspective of signal and filter, we compare one AR-RMDN with a modulation-spectrum-based post-filtering method that also eases the over-smoothing effect. It is demonstrated that the AR-RMDN also alters the modulation spectrum of the generated data trajectories but in a different way from the post-filtering method. The AR-RMDN also generates synthetic speech with better perceived quality. Based on the signal and filter interpretation, we further extend the AR-RMDN so that the inverse AR filter can acquire complex poles and stay stable.","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Neural-network-based mixture density networks are important tools for acoustic modeling in statistical parametric speech synthesis. Recently we found that incorporating an autoregressive model in a recurrent mixture density network, which is referred to as AR-RMDN, enabled the network to generate quite smooth acoustic data trajectories without using the delta and delta-delta coefficients. More interestingly, the new model generated trajectories with a dynamic range similar to that of the natural data, thus alleviating over-smoothing effect. In this work, after explaining the AR-RMDN from the perspective of signal and filter, we compare one AR-RMDN with a modulation-spectrum-based post-filtering method that also eases the over-smoothing effect. It is demonstrated that the AR-RMDN also alters the modulation spectrum of the generated data trajectories but in a different way from the post-filtering method. The AR-RMDN also generates synthetic speech with better perceived quality. Based on the signal and filter interpretation, we further extend the AR-RMDN so that the inverse AR filter can acquire complex poles and stay stable.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2017-02-10","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"2","bibliographicVolumeNumber":"2017-SLP-115"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"created":"2025-01-19T00:46:56.312612+00:00","id":177376,"links":{}}