Diverse and Controllable Speech Synthesis with GMM-Based Phone-Level Prosody Modelling.
| So they set out in a coach which was big enough to hold them, and had not gone very far when the hoodie suddenly said | |||
| Reference prosody | Target speaker | ||
| PLP-GMM-SP | |||
| fine-grained VAE | |||
| PLP-GMM-CP | |||
| The two men in red stopped on either side of this door. | |||
| Reference prosody | Target speaker | ||
| PLP-GMM-SP | |||
| fine-grained VAE | |||
| PLP-GMM-CP | |||
| I have wealth and power and knowledge above most men, but when the clock strikes I am afraid. | |||
| Reference prosody | Target speaker | ||
| PLP-GMM-SP | |||
| fine-grained VAE | |||
| PLP-GMM-CP | |||
| This makes it practically a note, to be paid at maturity. | |||
| Reference prosody | Target speaker | ||
| PLP-GMM-SP | |||
| fine-grained VAE | |||
| PLP-GMM-CP | |||