Pronunciation Accuracy Alignment Prediction on Mandarin Gold (test)
0.613LCCGemini-2.5-Pro
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gemini-2.5-ProInference Strategy=dimension-wise inference2026.04 | 0.613 | 0.53 | 0.048 | |
| TTS-PRISMParameters=7B, Inference Strategy=single-pass inference2026.04 | 0.511 | 0.492 | 0.073 | |
| Step-Audio-R1Parameters=33B, Inference Strategy=dimension-wise inference2026.04 | 0.475 | 0.423 | 0.081 | |
| Qwen3-OmniParameters=30B, Inference Strategy=dimension-wise inference2026.04 | 0.169 | 0.15 | 0.202 |