Target Speaker Extraction on REAL-T AliMeeting Chinese
2.086DNSMOS OVRLAlphaFlowTSE
Evaluation Results
| Method | Links | |
|---|---|---|
| AlphaFlowTSEMR predictor=with, Training set=Libri2Mix noisy, Zero-shot=true2026.03 | 2.086 | |
| MeanFlowTSEMR predictor=with, Training set=Libri2Mix noisy, Zero-shot=true2026.03 | 2.058 | |
| AlphaFlowTSEMR predictor=without, Training set=Libri2Mix noisy, Zero-shot=true2026.03 | 1.974 | |
| AD-FlowTSEMR predictor=with, Training set=Libri2Mix noisy, Zero-shot=true2026.03 | 1.855 | |
| AD-FlowTSEMR predictor=without, Training set=Libri2Mix noisy, Zero-shot=true2026.03 | 1.824 | |
| MeanFlowTSEMR predictor=without, Training set=Libri2Mix noisy, Zero-shot=true2026.03 | 1.607 |