Automatic Speech Recognition on GigaSpeech Thai 2
36.9Word Error Rate (WER)MiDashengLM
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MiDashengLMModel Size=7B2025.08 | 36.9 | 36.9 | |
| Qwen2.5-OmniModel Size=7B2025.08 | 53.8 | 53.8 | |
| Kimi-Audio-InstructModel Size=7B2025.08 | 100 | 100 |