Audio Question Answering on WenetSpeech-QA
48.47Audio-QA ScoreGLM-4-Voice
Evaluation Results
| Method | Links | |
|---|---|---|
| GLM-4-VoiceSize=9B2025.09 | 48.47 | |
| Kimi-AudioSize=7B2025.09 | 43.2 | |
| VITA-AudioSize=7B2025.09 | 30.75 | |
| LLaMA-OmniSize=8B2025.09 | 30.28 | |
| SpeechGPTSize=7B2025.09 | 24.53 | |
| Pretrain+TtTSize=3B2025.09 | 21.43 | |
| MoshiSize=7B2025.09 | 16.85 | |
| TtTSize=3B2025.09 | 11.61 | |
| SLAM-OmniSize=0.5B2025.09 | 7.9 | |
| Mini-OmniSize=0.5B2025.09 | 2.42 | |
| Qwen2.5-3B (AR)Size=3B2025.09 | 0.7 | |
| Qwen2.5-3B (NAR)Size=3B2025.09 | 0.7 |