Audio Clarity Alignment Prediction on Mandarin Gold (test)
0.815LCCTTS-PRISM
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TTS-PRISMParameters=7B, Inference Strategy=single-pass inference2026.04 | 0.815 | 0.826 | 0.018 | |
| Gemini-2.5-ProInference Strategy=dimension-wise inference2026.04 | 0.756 | 0.594 | 0.032 | |
| Step-Audio-R1Parameters=33B, Inference Strategy=dimension-wise inference2026.04 | 0.709 | 0.69 | 0.057 | |
| Qwen3-OmniParameters=30B, Inference Strategy=dimension-wise inference2026.04 | 0.665 | 0.685 | 0.065 |