Audio Understanding on X-Ares
99.81ASV2015SPEAR-XLarge
Evaluation Results
| Method | Links | ||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SPEAR-XLargeModel Type=Encoder Model2026.02 | 99.81 | — | — | — | — | — | — | — | — | — | — | — | 3.37 | 53.42 | 88.45 | 44.55 | 64.82 | 83.78 | — | — | — | — | |
| Dasheng2025.08 | 99.3 | 61.9 | 85.4 | 98.5 | 92.3 | 97.4 | 77 | 98.1 | 76.1 | 31.2 | 93.2 | 73.3 | 5.8 | 53.7 | 94.3 | 55.5 | 82.2 | 87.9 | 67.2 | 88.6 | 54.5 | 72.2 | |
| Dasheng 0.6BModel Type=Encoder Model2026.02 | 99.24 | — | — | — | — | — | — | — | — | — | — | — | 4 | 55.01 | 87.95 | 44.79 | 61.63 | 84.71 | — | — | — | — | |
| SPEAR-BaseModel Type=Encoder Model2026.02 | 99.2 | — | — | — | — | — | — | — | — | — | — | — | 3.65 | 48.63 | 82.25 | 37.96 | 48 | 81.46 | — | — | — | — | |
| MingTok-Audio (Unified)Model Type=Continuous Tokenizer2026.02 | 98.7 | — | — | — | — | — | — | — | — | — | — | — | 2 | 39.8 | 62.25 | 24.83 | 40.81 | 72.87 | — | — | — | — | |
| Whisper-Large-V3Model Type=Encoder Model2026.02 | 97.94 | — | — | — | — | — | — | — | — | — | — | — | 3.1 | 22.64 | 62.45 | 32 | 49.56 | 75.74 | — | — | — | — | |
| Whisper-Large V32025.08 | 97.9 | 64.4 | 90 | 94.9 | 97.4 | 97.7 | 71.3 | 97.8 | 68.5 | 29.3 | 91.5 | 24.8 | 3.1 | 22.6 | 62.5 | 32 | 49.6 | 75.7 | 58.9 | 71.8 | 0 | 63.5 | |
| OursModel Type=Continuous Tokenizer2026.02 | 96.45 | — | — | — | — | — | — | — | — | — | — | — | 5.64 | 55.4 | 96.4 | 59.98 | 86.81 | 83.8 | — | — | — | — | |
| Dasheng-BaseModel Type=Encoder Model2026.02 | 96.29 | — | — | — | — | — | — | — | — | — | — | — | 3.3 | 53.2 | 86.9 | 40.9 | 55.7 | 83.5 | — | — | — | — | |
| HuBERTModel Type=Encoder Model2026.02 | 96.24 | — | — | — | — | — | — | — | — | — | — | — | 1.74 | 3.64 | 37.35 | 17.4 | 26.5 | 57.4 | — | — | — | — | |
| WavLMModel Type=Encoder Model2026.02 | 95.37 | — | — | — | — | — | — | — | — | — | — | — | 0.79 | 16.76 | 31.45 | 10.06 | 19.63 | 53.13 | — | — | — | — | |
| EncodecModel Type=Codec2026.02 | 95.14 | — | — | — | — | — | — | — | — | — | — | — | 0.57 | 3.82 | 16.8 | 4.63 | 6.94 | 40.34 | — | — | — | — | |
| WavTokenizerModel Type=Codec2026.02 | 95 | — | — | — | — | — | — | — | — | — | — | — | 0 | 16 | 19 | 0 | 0 | 0 | — | — | — | — | |
| UniFlow-AudioModel Type=Continuous Tokenizer2026.02 | 93.77 | — | — | — | — | — | — | — | — | — | — | — | 0.57 | 16.9 | 19.85 | 4.83 | 12.19 | 35.16 | — | — | — | — | |
| SemantiCodecModel Type=Codec2026.02 | 93.69 | — | — | — | — | — | — | — | — | — | — | — | 1.89 | 48.57 | 81.75 | 34.84 | 26.69 | 84.54 | — | — | — | — | |
| DACModel Type=Codec2026.02 | 93.46 | — | — | — | — | — | — | — | — | — | — | — | 0.73 | 20.2 | 31.2 | 7.8 | 16.44 | 50.26 | — | — | — | — | |
| MSM-MAEModel Type=Encoder Model2026.02 | 93.19 | — | — | — | — | — | — | — | — | — | — | — | 3.86 | 0 | 88.7 | 44.15 | 48.63 | 84.35 | — | — | — | — | |
| FlexiCodecModel Type=Codec2026.02 | 93 | — | — | — | — | — | — | — | — | — | — | — | 0 | 17 | 30 | 0 | 0 | 0 | — | — | — | — | |
| MingTok-Audio (Acoustic)Model Type=Continuous Tokenizer2026.02 | 92.41 | — | — | — | — | — | — | — | — | — | — | — | 0.63 | 15.98 | 23.15 | 5.82 | 13.88 | 44.85 | — | — | — | — | |
| EzAudioModel Type=Continuous Tokenizer2026.02 | 91.44 | — | — | — | — | — | — | — | — | — | — | — | 0.4 | 17.01 | 18.05 | 4.86 | 13.38 | 34.97 | — | — | — | — | |
| OpenBeatsModel Type=Encoder Model2026.02 | 0 | — | — | — | — | — | — | — | — | — | — | — | 4 | 55.2 | 86.8 | 38 | 68.9 | 86.3 | — | — | — | — |