Speaker and Emotion Editing on ESD (test)
2.433CERGT
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GTSetting=Ground Truth2026.06 | 2.433 | 75.1 | 94.1 | 4.053 | 4.474 | 4.105 | |
| UniSAESetting=Unseen-Spk2026.06 | 8.017 | 60.5 | 70.3 | 3.684 | 2.105 | 3.579 | |
| ZESTSetting=Unseen-Spk2026.06 | 8.317 | 52.8 | 59.7 | 2.316 | 1.316 | 3.368 | |
| UniSAESetting=Seen-Spk2026.06 | 8.447 | 70.9 | 76.2 | 3.579 | 3.895 | 4.158 | |
| ZESTSetting=Seen-Spk2026.06 | 9.012 | 54.4 | 76.8 | 2.579 | 1.789 | 3.316 | |
| EmoConv-DiffSetting=Unseen-Spk2026.06 | 9.379 | 64.8 | 58 | 3.053 | 3.053 | 2.789 | |
| EmoConv-DiffSetting=Seen-Spk2026.06 | 10 | 68.3 | 62.5 | 3.105 | 2.789 | 2.316 |