WEKO3
アイテム
Eigenvoice-Based Approach to Voice Conversion and Voice Quality Control
http://hdl.handle.net/10061/8235
http://hdl.handle.net/10061/8235f12d8c07-c897-42dc-93e0-12aa40c07275
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
fulltext (194.6 kB)
|
|
| アイテムタイプ | 会議発表論文 / Conference Paper(1) | |||||
|---|---|---|---|---|---|---|
| 公開日 | 2012-08-22 | |||||
| タイトル | ||||||
| タイトル | Eigenvoice-Based Approach to Voice Conversion and Voice Quality Control | |||||
| 言語 | ||||||
| 言語 | eng | |||||
| 資源タイプ | ||||||
| 資源タイプ | conference paper | |||||
| アクセス権 | ||||||
| アクセス権 | open access | |||||
| 著者 |
Toda, Tomoki
× Toda, Tomoki |
|||||
| 抄録 | ||||||
| 内容記述タイプ | Abstract | |||||
| 内容記述 | This paper reviews our proposed approach to voice conversion (VC) and voice quality control based on an eigenvoice technique. VC is a technique to modify nonlinguistic information such as speaker individuality while keeping linguistic information unchanged. In the traditional VC framework, a conversion model for a source and target speaker-pair needs to be trained in advance using a parallel data set consisting of utterance-pairs of these two speakers. To make VC technologies more practical, we have developed a new VC paradigm for flexibly building the conversion model for an arbitrary speaker-pair by effectively using speech samples of many other speakers. In this paper, we give an overview of eigenvoice conversion (EVC) as one of our proposed VC techniques. | |||||
| 書誌情報 |
p. 492-497, 発行日 2009-08 |
|||||
| 会議情報 | ||||||
| 会議名 | NCMMSC2009: National Conference on Man-Machine Speech Communication | |||||
| 開催期間 | August 14-19, 2009 | |||||
| 開催地 | Lanzhou Gansu Province | |||||
| 開催国 | CHN | |||||
| 著者版フラグ | ||||||
| 出版タイプ | VoR | |||||
