| 講演抄録/キーワード |
| 講演名 |
2013-12-19 16:00
Improvement of AF-based Voice Conversion for Arbitrary Speakers ○Narpendyah Wisjnu Ariwardhani(TUT)・Yurie Iribe(Aichi Prefectureal Univ.)・Kouichi Katsurada(TUT)・Tsuneo Nitta(Waseda Univ.) SP2013-85 |
| 抄録 |
(和) |
In this paper, we use artificial neural networks (ANNs) for articulatory feature (AF) based voice conversion. ANNs are applied to map AF to vocal tract parameter (VTP) and to convert the source speaker?s voice to the target speaker?s voice. The proposed system is not only text independent voice conversion, but can also be used for an arbitrary source speaker. This means that our approach requires no source speaker data to build the voice conversion model and hence source speaker data is only required during testing phase. The results of voice conversion were evaluated using subjective and objective measures to compare the performance of our proposed ANN-based voice conversion (VC) with the state-of-the-art Gaussian mixture model (GMM)-based VC. The experimental results show that the converted voice is intelligible and has speaker individuality of the target speaker. Furthermore, preliminary experiments are also conducted to investigate AF flexibility on cross-lingual voice conversion. |
| (英) |
In this paper, we use artificial neural networks (ANNs) for articulatory feature (AF) based voice conversion. ANNs are applied to map AF to vocal tract parameter (VTP) and to convert the source speaker?s voice to the target speaker?s voice. The proposed system is not only text independent voice conversion, but can also be used for an arbitrary source speaker. This means that our approach requires no source speaker data to build the voice conversion model and hence source speaker data is only required during testing phase. The results of voice conversion were evaluated using subjective and objective measures to compare the performance of our proposed ANN-based voice conversion (VC) with the state-of-the-art Gaussian mixture model (GMM)-based VC. The experimental results show that the converted voice is intelligible and has speaker individuality of the target speaker. Furthermore, preliminary experiments are also conducted to investigate AF flexibility on cross-lingual voice conversion. |
| キーワード |
(和) |
voice conversion / neural network / articulatory feature / arbitrary speaker / cross-lingual / / / |
| (英) |
voice conversion / neural network / articulatory feature / arbitrary speaker / cross-lingual / / / |
| 文献情報 |
信学技報, vol. 113, no. 366, SP2013-85, pp. 65-70, 2013年12月. |
| 資料番号 |
SP2013-85 |
| 発行日 |
2013-12-12 (SP) |
| ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
| PDFダウンロード |
SP2013-85 |