Multi-speaker dialog synthesis on COVOMIX2-DIALOGUE-20S
0.145cpWERSCENA
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| SCENA2026.06 | 0.145 | 0.567 | 0.866 | 0.02 | 0.451 | 3.44 | 4.32 | |
| ZipVoice-Dialog2026.06 | 0.176 | 0.538 | 0.847 | 0.032 | 0.446 | 3.57 | 4.34 | |
| VibeVoice-7BModel size=7B2026.06 | 0.206 | 0.527 | 0.821 | 0.044 | 0.451 | 3.58 | 4.28 | |
| VibeVoice-1.5BModel size=1.5B2026.06 | 0.212 | 0.503 | 0.83 | 0.05 | 0.423 | 3.56 | 4.27 | |
| MOSS-TTSD2026.06 | 0.232 | 0.547 | 0.855 | 0.109 | 0.443 | 3.76 | 4.28 | |
| Dia (Nari Labs)2026.06 | 0.303 | 0.339 | 0.757 | 0.133 | 0.312 | 2.69 | 4.09 |