Classification accuracy (%) for text representations from WER-358 noisy models. Classification 1: Clean-style vs Noisy-style text. Classification 2: Four SNR-based quality levels
| Models | Classification 1 | Classification 2 |
|---|---|---|
| Tacotron2 | 61.1 | 36.1 |
| Vits-dp | 81.5 | 55.1 |
| GradTTS | 73.5 | 41.0 |
| Dvt | 83.9 | 49.1 |
| Dvt2 | 85.7 | 54.1 |
| Models | Classification 1 | Classification 2 |
|---|---|---|
| Tacotron2 | 61.1 | 36.1 |
| Vits-dp | 81.5 | |
| GradTTS | 73.5 | 41.0 |
| Dvt | 83.9 | 49.1 |
| Dvt2 | 54.1 |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.