Question Answering on SDQA
63.1AccuracyKimi-Audio
Evaluation Results
| Method | Links | |
|---|---|---|
| Kimi-AudioFull Duplex=false2026.07 | 63.1 | |
| Freeze-OmniFull Duplex=true2026.07 | 53.5 | |
| DuplexPOFull Duplex=true2026.07 | 49.8 | |
| SFT BaselineFull Duplex=true2026.07 | 47.2 | |
| Baichuan-AudioFull Duplex=false2026.07 | 45.8 | |
| APinSetting=Aggressive (τin=0.80, τdeep=0.70), lin=✓, ldeep=-, FRR=86.28, Pre. GFLOPs=351.18, FLOPs Ratio=86.242026.04 | 37.79 | |
| APdeepSetting=Aggressive (τin=0.80, τdeep=0.70), lin=-, ldeep=✓, FRR=48.81, Pre. GFLOPs=377.39, FLOPs Ratio=92.682026.04 | 37.79 | |
| DAPSetting=Aggressive (τin=0.80, τdeep=0.70), lin=✓, ldeep=✓, FRR=40.71, Pre. GFLOPs=324.64, FLOPs Ratio=79.722026.04 | 37.61 | |
| APdeepSetting=Conservative (τin=0.90, τdeep=0.80), lin=-, ldeep=✓, FRR=71.18, Pre. GFLOPs=390.42, FLOPs Ratio=95.872026.04 | 37.61 | |
| GLM-4-VoiceFull Duplex=false2026.07 | 37 | |
| DAPSetting=Conservative (τin=0.90, τdeep=0.80), lin=✓, ldeep=✓, FRR=68.19, Pre. GFLOPs=376.14, FLOPs Ratio=92.362026.04 | 36.89 | |
| Vanillalin=-, ldeep=-, FRR=100.0, Pre. GFLOPs=407.22, FLOPs Ratio=100.02026.04 | 36.71 | |
| APinSetting=Conservative (τin=0.90, τdeep=0.80), lin=✓, ldeep=-, FRR=96.44, Pre. GFLOPs=392.69, FLOPs Ratio=96.432026.04 | 36.71 | |
| Qwen2-AudioFull Duplex=false2026.07 | 35.7 | |
| DAPlin=✓, ldeep=✓, FRR=14.91, Pre. GFLOPs=566.30, FLOPs Ratio=72.52, Setting=Aggressive (τin=0.80, τdeep=0.70)2026.04 | 29.66 | |
| DAPlin=✓, ldeep=✓, FRR=33.76, Pre. GFLOPs=686.39, FLOPs Ratio=87.89, Setting=Conservative (τin=0.90, τdeep=0.80)2026.04 | 27.85 | |
| APinlin=✓, ldeep=-, FRR=93.56, Pre. GFLOPs=730.28, FLOPs Ratio=93.51, Setting=Conservative (τin=0.90, τdeep=0.80)2026.04 | 27.67 | |
| APdeeplin=-, ldeep=✓, FRR=14.30, Pre. GFLOPs=718.12, FLOPs Ratio=91.96, Setting=Aggressive (τin=0.80, τdeep=0.70)2026.04 | 27.49 | |
| Vanillalin=-, ldeep=-, FRR=100.0, Pre. GFLOPs=780.94, FLOPs Ratio=100.0, Setting=N/A2026.04 | 27.31 | |
| APdeeplin=-, ldeep=✓, FRR=33.29, Pre. GFLOPs=731.96, FLOPs Ratio=93.73, Setting=Conservative (τin=0.90, τdeep=0.80)2026.04 | 26.94 | |
| APinlin=✓, ldeep=-, FRR=78.64, Pre. GFLOPs=612.93, FLOPs Ratio=78.49, Setting=Aggressive (τin=0.80, τdeep=0.70)2026.04 | 26.29 | |
| SALM-DuplexFull Duplex=true2026.07 | 26 | |
| MoshiFull Duplex=true2026.07 | 15.6 |