TDANet Large

Reported on 2 benchmarks across 1 task · 1 paper · 2 SOTA

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

Speech2 results

Speech SeparationonLibri2Mix
SI-SDRi· 2022-09-30
17.4
best: 22.2 (MossFormer2 (w speed perturb))
SOTA
An efficient encoder-decoder architecture with top-down attention for speech separation arXiv:2209.15200
Speech SeparationonWHAM!
SI-SDRi· 2022-09-30
15.2
best: 18.4 (SepReformer-L + DM)
SOTA
An efficient encoder-decoder architecture with top-down attention for speech separation arXiv:2209.15200