Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Noisy Speech Separation on LibriSpeech + FUSS 2 Source Mixtures (test)
Loading...
18.21
SI-SDR
Supervised Flow
14.0188
15.1069
16.195
17.2831
Jun 3, 2026
SI-SDR
ESTOI
PESQ
DNSMOS
Updated 1mo ago
Evaluation Results
Method
Method
Links
SI-SDR
ESTOI
PESQ
DNSMOS
Supervised Flow
Mode=Supervised
2026.06
18.21
0.904
3.29
2.58
SURF
Teacher=Self-Remixing,...
2026.06
15.23
0.84
2.93
2.57
SURF
Teacher=ReMixIT, Mode=...
2026.06
14.98
0.84
2.86
2.55
Self-Remixing
Mode=Unsupervised
2026.06
14.81
0.784
2.81
2.56
ReMixIT
Mode=Unsupervised
2026.06
14.29
0.793
2.85
2.51
MixIT
Mode=Unsupervised
2026.06
14.18
0.771
2.79
2.53
Feedback
Search any
task
Search any
task