Colab inference for ZFTurbo's Music-Source-Separation-Training
New notebook added to ease the use of custom models ! NEW !
You can find a list of models here (maintained by Bas Curtiz):
https://bascurtiz.x10.mx/models-checkpoint-config-urls.html or
https://github.com/SiftedSand/MusicSepGUI/blob/main/models.json (by SeinfeldMaster)
Available models:
"INST-MelBand-Roformer INSTV7 (by Gabox)" NEW !
"INST-MelBand-Roformer INSTFVX (by Gabox)" NEW !
"INST-MelBand-Roformer INSTV7N (by Gabox)" NEW !
"INST-Mel-Roformer gabox_inst3 (by Gabox)" NEW !
"INST-Mel-Roformer Kim FT2 bleedless (by unwa)" NEW !
"VOCALS-MelBand-Roformer Big Beta 6 (by Unwa)"
"INST-MelBand-Roformer INSTV6N (by Gabox)"
"VOCALS-MelBand-Roformer voc_Fv4 (by Gabox)"
"VOCALS-MelBand-Roformer voc_Fv3 (by Gabox)"
"INST-Mel-Roformer INSTV6 (by Gabox)"
"INST-Mel-Roformer INSTV5 (by Gabox)
"VOCALS-Male-Female-Mel-Roformer 7.2889 FT (by Aufr33)"
"VOCALS-Mel-Roformer Kim FT 2 (by unwa)"
"INST-Mel-Roformer (by Becruily)"
"VOCAL-Mel-Roformer (by Becruily)"
"DEBLEED-Mel-Roformer (by unwa/97chris)"
"4STEMS-SCNet-XL-MUSDB18 (by ZFTurbo)"
"4STEMS-BS-Roformer-MUSDB18 (by ZFTurbo)"
"4STEMS-SCNet-large (by starrytong)"
"VOCALS-Mel-Roformer Kim FT (by unwa)"
"4STEMS-SCNet-MUSDB18 (by starrytong)"
"VOCALS-Mel-Roformer BigBeta5e (by unwa)"
"INST-Mel-Roformer v1e (by unwa)"
"INST-Mel-Roformer v2 (by unwa)"
"INST-VOC-Mel-Roformer a.k.a. duality v2 (by unwa)"
"INST-Mel-Roformer v1 (by unwa)"
"INST-VOC-Mel-Roformer a.k.a. duality (by unwa)"
"VOCALS-Mel-Roformer big beta 4 (by unwa)"
"BS-Roformer Large V1 (by unwa)"
"MDX23C DeReverb (by aufr33 and jarredou)"
"MelBand-Roformer Karaoke (by aufr33 and viperx)"
"Kim MelBand-vocals" (by KimberleyJSN)
"MelBand-Denoise" (by aufr33)
"MDX23C_InstVocHQ" (by Anjok, aufr33, ZFTurbo)
"BS-Roformer_1297" (by viperx)
"BS-Roformer_1296" (by viperx)
"BS-Roformer_1053" (by viperx)
"SCNet_MUSDB18" (by starrytong)
"MelBand-Roformer_Decrowd" (by aufr33)
"VitLarge23" (by ZFTurbo)
"BandIt_Plus" (by kwatcharasupat)
"MDX23C_DrumSep_6stem" (by jarredou & aufr33)
For becruily instrumental model consider using this Colab:
https://colab.research.google.com/github/lucassantillifuck2fa/Music-Source-Separation-Training/blob/main/Phase_Fixer.ipynb with auto phase fix for results with less vocal residues.