RemixIT (w Sudo U=32)

Reported on 2 benchmarks across 1 task · 2 papers

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio4 results

Speech EnhancementonDeep Noise Suppression (DNS) Challenge
PESQ-WB· uses extra data· 2022-02-17
2.34
best: 3.81 (ZipEnhancer (M))
RemixIT: Continual self-training of speech enhancement models via bootstrapped remixing arXiv:2202.08862
Speech EnhancementonDeep Noise Suppression (DNS) Challenge
SI-SDR-WB· uses extra data· 2022-02-17
16
best: 23.3 (TF-Locoformer (M))
RemixIT: Continual self-training of speech enhancement models via bootstrapped remixing arXiv:2202.08862
Speech EnhancementonDeep Noise Suppression (DNS) Challenge
PESQ-WB· uses extra data· 2021-10-19
2.6
best: 3.81 (ZipEnhancer (M))
Continual self-training with bootstrapped remixing for speech enhancement arXiv:2110.10103
Speech EnhancementonDeep Noise Suppression (DNS) Challenge
SI-SDR-WB· uses extra data· 2021-10-19
18
best: 23.3 (TF-Locoformer (M))
Continual self-training with bootstrapped remixing for speech enhancement arXiv:2110.10103