Automatic Speech Recognition on Fen
4.87WERKimi-Audio
Evaluation Results
| Method | Links | |
|---|---|---|
| Kimi-AudioSize=7B2025.09 | 4.87 | |
| VITA-AudioSize=7B2025.09 | 9.58 | |
| Pretrain+TtTSize=3B2025.09 | 19.1 | |
| Mini-OmniSize=0.5B2025.09 | 22.74 | |
| SpeechGPTSize=7B2025.09 | 45.15 | |
| TtTSize=3B2025.09 | 64.31 | |
| Qwen2.5-3B (AR)Size=3B2025.09 | 74.47 | |
| Qwen2.5-3B (NAR)Size=3B2025.09 | 83.51 |