DC Field | Value | Language |
---|---|---|
dc.contributor.author | Park, Sangjun | ko |
dc.contributor.author | Hahn, Minsoo | ko |
dc.date.accessioned | 2018-12-20T06:52:44Z | - |
dc.date.available | 2018-12-20T06:52:44Z | - |
dc.date.created | 2018-11-30 | - |
dc.date.created | 2018-11-30 | - |
dc.date.issued | 2018-09 | - |
dc.identifier.citation | JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA, v.144, no.3, pp.EL151 - EL157 | - |
dc.identifier.issn | 0001-4966 | - |
dc.identifier.uri | http://hdl.handle.net/10203/248313 | - |
dc.description.abstract | In statistical parametric speech synthesis, a mixture density network is employed to address the limitations of a linear output layer such as pre-computed fixed variances and the unimodal assumption. However, it also has a defect, i.e., it cannot deploy a static-dynamic constraint needed in the training phase for high-quality speech synthesis. To cope with this problem, this paper proposes a training algorithm based on the minimum trajectory error for a mixture density network. And a modulation spectrum-constrained loss function is also proposed to alleviate the over-smoothing effect. The experimental results confirm meaningful improvement both in objective and subjective performance measures. | - |
dc.language | English | - |
dc.publisher | ACOUSTICAL SOC AMER AMER INST PHYSICS | - |
dc.title | Modulation spectrum-constrained trajectory error training for mixture density network-based speech synthesis | - |
dc.type | Article | - |
dc.identifier.wosid | 000457802200001 | - |
dc.identifier.scopusid | 2-s2.0-85052990891 | - |
dc.type.rims | ART | - |
dc.citation.volume | 144 | - |
dc.citation.issue | 3 | - |
dc.citation.beginningpage | EL151 | - |
dc.citation.endingpage | EL157 | - |
dc.citation.publicationname | JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA | - |
dc.identifier.doi | 10.1121/1.5052206 | - |
dc.contributor.localauthor | Hahn, Minsoo | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.