Spoken Dialogue Generation on TTSD EN (test)
96.55AccuracyMOSS-TTSD
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MOSS-TTSDReference Speaker Timbre=Gemini2026.03 | 96.55 | 78.93 | 9.84 | |
| MOSS-TTSDContext=Open-Source Models comparison2026.03 | 96.26 | 73.26 | 9.88 | |
| MOSS-TTSDReference Speaker Timbre=ElevenLabs2026.03 | 95.65 | 73.04 | 10.05 | |
| VibeVoiceModel Size=7B2026.03 | 95.54 | 71.4 | 9.46 | |
| gemini-2.5-pro-preview-tts2026.03 | 95.37 | 67.86 | 8.59 | |
| gemini-2.5-flash-preview-tts2026.03 | 95.11 | 71.94 | 8.71 | |
| Eleven V32026.03 | 94.98 | 67.3 | 8.24 | |
| VibeVoiceModel Size=1.5B2026.03 | 93.53 | 69.61 | 11.33 | |
| Higgs Audio V22026.03 | 90.25 | 68.6 | 21.31 |