461 |
4081 |
30 Best Models |
? |
+ |
--- |
--- |
--- |
9.5114 |
15.7502 |
462 |
5142 |
UVR-MDX-NET-Voc_FT (segment size: 256 | overlap: default | denoise model) |
? |
- |
--- |
--- |
--- |
9.5086 |
15.8160 |
463 |
3787 |
ENSEMBLE | Voc FT + HQ3 (as secondary model 20%) | Avg/Avg | Chunks: ON |
? |
- |
--- |
--- |
--- |
9.5072 |
15.7319 |
464 |
3780 |
MDX | Voc FT | Chunks: OFF | Volume Compensation: AUTO | Denoise: ON | Spectral inversion: OFF |
? |
- |
--- |
--- |
--- |
9.5028 |
15.7781 |
465 |
1662 |
Ensemble | MDX'23 Model + Kim Vocal 1 + Kim Vocal 2 + Main 427 | Volume Compensation: Auto | Max/Max | Chunks Batch |
? |
+ |
--- |
--- |
--- |
9.5020 |
15.7039 |
466 |
1214 |
Ensemble | Kim Inst + Inst HQ 1 + Inst HQ 2 + htdemucs_Ft | Avg/Avg |
? |
+ |
--- |
--- |
--- |
9.4920 |
15.6796 |
467 |
4237 |
MDX23C, 12K FFT, Full Band, 6 Poolings (MVSep.com)
|
? |
- |
--- |
--- |
--- |
9.4917 |
15.7994 |
468 |
993 |
Ensemble | MDX23 Best Model + Kim vocal 1 + Kim FT other + Main 427 + htdemucs_ft | Avg/Avg | Chunks Batch | Denoise Output ON | Spectral Inversion OF |
? |
+ |
--- |
--- |
--- |
9.4906 |
15.7944 |
469 |
4367 |
UVR-MDX-NET Voc_FT - A5 Settings (stereo processed) |
? |
- |
--- |
--- |
--- |
9.4902 |
15.7557 |
470 |
5115 |
UVR-MDX-NET-Voc_FT (segment size: 256 | overlap: default) |
? |
- |
--- |
--- |
--- |
9.4902 |
15.7977 |
471 |
409 |
ENSEMBLE | Kim_vocal_1 + UVR-MDX-NET Inst Main + Demucs v4 htdemucs_ft | Avg/Avg |
? |
+ |
--- |
--- |
--- |
9.4882 |
15.7879 |
472 |
3782 |
MDX | Kim Vocal 2 | Chunks: ON | Volume Compensation: AUTO | Denoise: ON | Spectral inversion: OFF |
? |
- |
--- |
--- |
--- |
9.4781 |
15.7378 |
473 |
475 |
Ensemble | UVR-MDX-NET-Inst HQ 1 + Kim_Vocal_1 | Chunks Disabled | Volume Compensation 1.035 | Denoise Output ON | Spectral Inversion OFF | WAV |
? |
+ |
--- |
--- |
--- |
9.4629 |
15.7270 |
474 |
6906 |
Kim_Vocal_2 on MPS | { "hop_length": 1024,"segment_size": 256,"overlap": 0.25,"batch_size": 1,"enable_denoise": False } |
? |
- |
--- |
--- |
--- |
9.4610 |
15.7677 |
475 |
1661 |
Ensemble | MDX'23 Model + Kim Vocal 1 + Kim Vocal 2 + Main 427 | Volume Compensation: Auto | Avg/Avg | Chunks Batch |
? |
+ |
--- |
--- |
--- |
9.4430 |
15.7527 |
476 |
928 |
Kim vocal 2 | Batch Mode |
? |
- |
--- |
--- |
--- |
9.4399 |
15.6424 |
477 |
2500 |
MDX-Colab Kim Vocal 2 no_lp |
? |
- |
--- |
--- |
--- |
9.4380 |
15.7455 |
478 |
322 |
MDX | Kim_Vocal_1 | Chunks AUTO | Margin Size 44100 | Volume Compensation 1.035 | Denoise Output ON | Spectral Inversion OFF |
? |
- |
--- |
--- |
--- |
9.3660 |
15.7069 |
479 |
395 |
MDX | Kim_Vocal_1 | Chunks AUTO | Volume Compensation 1.08 | Denoise Output ON | Spectral Inversion OFF |
? |
- |
--- |
--- |
--- |
9.3660 |
15.6281 |
480 |
396 |
MDX | Kim_Vocal_1 | Chunks AUTO | Volume Compensation AUTO | Denoise Output ON | Spectral Inversion OFF |
? |
- |
--- |
--- |
--- |
9.3660 |
15.6448 |