SepMamba: State-space models for speaker separation using Mamba
Open paper
Benchmark
Model
Rank
Results
speech-separation-on-wsj0-2mix
SepMamba + DM (M)
#10
SI-SDRi: 22.7
SDRi: 22.9
speech-separation-on-wsj0-2mix
SepMamba + DM (S)
#17
SI-SDRi: 21.2
SDRi: 21.4
Rank counts only results with a code link.