Spoken Language Modeling on Libri-Light 6k
71.89sWUGGY (all)SpidR
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| SpidRLayer=6, Units=K-means, Zero-shot protocol=true2025.12 | 71.89 | 82.46 | 59.48 | 70.46 | |
| SpidRLayer=6, Units=Codebook, Zero-shot protocol=true2025.12 | 69.78 | 79.98 | 58.1 | 70.14 | |
| WavLM BaseLayer=11, Units=K-means, Zero-shot protocol=true2025.12 | 69.74 | 79.88 | 56.6 | 70.35 | |
| HuBERTLayer=11, Units=K-means, Zero-shot protocol=true2025.12 | 65.5 | 73.67 | 55.6 | 68.75 | |
| wav2vec 2.0Layer=6, Units=K-means, Zero-shot protocol=true2025.12 | 62.29 | 68.5 | 53.34 | 65.97 | |
| DinoSRLayer=5, Units=Codebook, Zero-shot protocol=true2025.12 | 60.1 | 64.56 | 57.04 | 69.44 | |
| DinoSRLayer=5, Units=Codebook, Implementation=re-implementation, Zero-shot protocol=true2025.12 | 58.93 | 62.07 | 55.79 | 66.08 | |
| DinoSRLayer=5, Units=K-means, Zero-shot protocol=true2025.12 | 56.69 | 59.42 | 54.32 | 65.76 | |
| DinoSRLayer=5, Units=K-means, Implementation=re-implementation, Zero-shot protocol=true2025.12 | 55.73 | 57.58 | 54.04 | 63.68 |