Single Speaker (LJSpeech Dataset)
| Utterance | groundthruth | UDPNet(fsteps:1200 rsteps 8) | UDPNet(fsteps:960 rsteps 8) | UDPNet(fsteps:720 rsteps 8) | UDPNet(fsteps:240 rsteps 8) |
|---|---|---|---|---|---|
| #1 | |||||
| #2 | |||||
| #3 | |||||
| #4 | |||||
| #5 |
Unseen Speakers (VCTK Dataset)
| Utterance | groundthruth | UDPNet(fsteps:1200 rsteps 8) | UDPNet(fsteps:960 rsteps 8) | UDPNet(fsteps:720 rsteps 8) | UDPNet(fsteps:240 rsteps 8) |
|---|---|---|---|---|---|
| #1 | |||||
| #2 | |||||
| #3 | |||||
| #4 | |||||
| #5 |