| Presentation # | 1 |
| Session: | Voice Conversion and TTS |
| Session Time: | Friday, December 21, 10:00 - 12:00 |
| Presentation Time: | Friday, December 21, 10:00 - 12:00 |
| Presentation: |
Poster
|
| Topic: |
Speech recognition and synthesis: |
| Paper Title: |
StarGAN-VC: Non-parallel many-to-many voice conversion using star generative adversarial networks |
| Authors: |
Hirokazu Kameoka; NTT Corporation | | |
| | Takuhiro Kaneko; NTT Corporation | | |
| | Kou Tanaka; NTT Corporation | | |
| | Nobukatsu Hojo; NTT Corporation | | |
| Abstract: |
This paper proposes a method that allows non-parallel many-to-many voice conversion (VC) by using a variant of a generative adversarial network (GAN) called StarGAN. Our method, which we call StarGAN-VC, is noteworthy in that it (1) requires no parallel utterances, transcriptions, or time alignment procedures for speech generator training, (2) simultaneously learns many-to-many mappings across different attribute domains using a single generator network, (3) is able to generate converted speech signals quickly enough to allow real-time implementations and (4) requires only several minutes of training examples to generate reasonably realistic sounding speech. Subjective evaluation experiments on a non-parallel many-to-many speaker identity conversion task revealed that the proposed method obtained higher sound quality and speaker similarity than a state-of-the-art method based on variational autoencoding GANs. |