forked from Anjok07/ultimatevocalremovergui
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathinference_v5.py
1123 lines (968 loc) · 62.4 KB
/
inference_v5.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
from collections import defaultdict
from datetime import datetime
from demucs.apply import BagOfModels, apply_model
from demucs.hdemucs import HDemucs
from demucs.pretrained import get_model as _gm
from lib_v5 import dataset
from lib_v5 import spec_utils
from lib_v5.model_param_init import ModelParameters
from models import stft, istft
from pathlib import Path
from random import randrange
from tqdm import tqdm
from tkinter import filedialog
import lib_v5.filelist
import cv2
import hashlib
import importlib
import librosa
import math
import numpy as np
import os
import pydub
import shutil
import soundfile as sf
import time # Timer
import tkinter as tk
import torch
import traceback # Error Message Recent Calls
class VocalRemover(object):
def __init__(self, data, text_widget: tk.Text):
self.data = data
self.text_widget = text_widget
self.models = defaultdict(lambda: None)
self.devices = defaultdict(lambda: None)
# self.offset = model.offset
data = {
'agg': 10,
'demucsmodel_sel_VR': 'UVR_Demucs_Model_1',
'demucsmodelVR': True,
'export_path': None,
'gpu': -1,
'high_end_process': 'mirroring',
'input_paths': None,
'inst_only': False,
'instrumentalModel': None,
'ModelParams': 'Auto',
'mp3bit': '320k',
'normalize': False,
'output_image': True,
'overlap': 0.5,
'postprocess': True,
'saveFormat': 'wav',
'segment': 'None',
'settest': False,
'shifts': 0,
'split_mode': False,
'tta': True,
'useModel': None,
'voc_only': False,
'wavtype': 'PCM_16',
'window_size': 512,
}
default_window_size = data['window_size']
default_agg = data['agg']
def update_progress(progress_var, total_files, file_num, step: float = 1):
"""Calculate the progress for the progress widget in the GUI"""
base = (100 / total_files)
progress = base * (file_num - 1)
progress += base * step
progress_var.set(progress)
def get_baseText(total_files, file_num):
"""Create the base text for the command widget"""
text = 'File {file_num}/{total_files} '.format(file_num=file_num,
total_files=total_files)
return text
def determineModelFolderName():
"""
Determine the name that is used for the folder and appended
to the back of the music files
"""
modelFolderName = ''
if not data['modelFolder']:
# Model Test Mode not selected
return modelFolderName
# -Instrumental-
if os.path.isfile(data['instrumentalModel']):
modelFolderName += os.path.splitext(os.path.basename(data['instrumentalModel']))[0]
if modelFolderName:
modelFolderName = '/' + modelFolderName
return modelFolderName
def main(window: tk.Wm, text_widget: tk.Text, button_widget: tk.Button, progress_var: tk.Variable,
**kwargs: dict):
global model_params_d
global nn_arch_sizes
global nn_architecture
global overlap_set
global shift_set
global split_mode
global demucs_model_set
global wav_type_set
global flac_type_set
global mp3_bit_set
wav_type_set = data['wavtype']
#Error Handling
runtimeerr = "CUDNN error executing cudnnSetTensorNdDescriptor"
systemmemerr = "DefaultCPUAllocator: not enough memory"
cuda_err = "CUDA out of memory"
mod_err = "ModuleNotFoundError"
file_err = "FileNotFoundError"
ffmp_err = """audioread\__init__.py", line 116, in audio_open"""
sf_write_err = "sf.write"
try:
with open('errorlog.txt', 'w') as f:
f.write(f'No errors to report at this time.' + f'\n\nLast Process Method Used: VR Architecture' +
f'\nLast Conversion Time Stamp: [{datetime.now().strftime("%Y-%m-%d %H:%M:%S")}]\n')
except:
pass
nn_arch_sizes = [
31191, # default
33966, 123821, 123812, 129605, 537238 # custom
]
nn_architecture = list('{}KB'.format(s) for s in nn_arch_sizes)
def save_files(wav_instrument, wav_vocals):
"""Save output music files"""
vocal_name = '(Vocals)'
instrumental_name = '(Instrumental)'
save_path = os.path.dirname(base_name)
# Swap names if vocal model
VModel="Vocal"
if VModel in model_name:
# Reverse names
vocal_name, instrumental_name = instrumental_name, vocal_name
# Save Temp File
# For instrumental the instrumental is the temp file
# and for vocal the instrumental is the temp file due
# to reversement
if data['demucsmodelVR']:
samplerate = 44100
else:
samplerate = mp.param['sr']
sf.write(f'temp.wav',
normalization_set(wav_instrument).T, samplerate, subtype=wav_type_set)
appendModelFolderName = modelFolderName.replace('/', '_')
# -Save files-
# Instrumental
if instrumental_name is not None:
if data['modelFolder']:
instrumental_path = '{save_path}/{file_name}.wav'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}{appendModelFolderName}_{instrumental_name}',)
instrumental_path_mp3 = '{save_path}/{file_name}.mp3'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}{appendModelFolderName}_{instrumental_name}',)
instrumental_path_flac = '{save_path}/{file_name}.flac'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}{appendModelFolderName}_{instrumental_name}',)
else:
instrumental_path = '{save_path}/{file_name}.wav'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}_{instrumental_name}',)
instrumental_path_mp3 = '{save_path}/{file_name}.mp3'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}_{instrumental_name}',)
instrumental_path_flac = '{save_path}/{file_name}.flac'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}_{instrumental_name}',)
if os.path.isfile(instrumental_path):
file_exists_i = 'there'
else:
file_exists_i = 'not_there'
if VModel in model_name and data['voc_only']:
sf.write(instrumental_path,
normalization_set(wav_instrument).T, samplerate, subtype=wav_type_set)
elif VModel in model_name and data['inst_only']:
pass
elif data['voc_only']:
pass
else:
sf.write(instrumental_path,
normalization_set(wav_instrument).T, samplerate, subtype=wav_type_set)
# Vocal
if vocal_name is not None:
if data['modelFolder']:
vocal_path = '{save_path}/{file_name}.wav'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}{appendModelFolderName}_{vocal_name}',)
vocal_path_mp3 = '{save_path}/{file_name}.mp3'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}{appendModelFolderName}_{vocal_name}',)
vocal_path_flac = '{save_path}/{file_name}.flac'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}{appendModelFolderName}_{vocal_name}',)
else:
vocal_path = '{save_path}/{file_name}.wav'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}_{vocal_name}',)
vocal_path_mp3 = '{save_path}/{file_name}.mp3'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}_{vocal_name}',)
vocal_path_flac = '{save_path}/{file_name}.flac'.format(
save_path=save_path,
file_name=f'{os.path.basename(base_name)}_{vocal_name}',)
if os.path.isfile(vocal_path):
file_exists_v = 'there'
else:
file_exists_v = 'not_there'
if VModel in model_name and data['inst_only']:
sf.write(vocal_path,
normalization_set(wav_vocals).T, samplerate, subtype=wav_type_set)
elif VModel in model_name and data['voc_only']:
pass
elif data['inst_only']:
pass
else:
sf.write(vocal_path,
normalization_set(wav_vocals).T, samplerate, subtype=wav_type_set)
if data['saveFormat'] == 'Mp3':
try:
if data['inst_only'] == True:
pass
else:
musfile = pydub.AudioSegment.from_wav(vocal_path)
musfile.export(vocal_path_mp3, format="mp3", bitrate=mp3_bit_set)
if file_exists_v == 'there':
pass
else:
try:
os.remove(vocal_path)
except:
pass
if data['voc_only'] == True:
pass
else:
musfile = pydub.AudioSegment.from_wav(instrumental_path)
musfile.export(instrumental_path_mp3, format="mp3", bitrate=mp3_bit_set)
if file_exists_i == 'there':
pass
else:
try:
os.remove(instrumental_path)
except:
pass
except Exception as e:
traceback_text = ''.join(traceback.format_tb(e.__traceback__))
errmessage = f'Traceback Error: "{traceback_text}"\n{type(e).__name__}: "{e}"\n'
if "ffmpeg" in errmessage:
text_widget.write(base_text + 'Failed to save output(s) as Mp3(s).\n')
text_widget.write(base_text + 'FFmpeg might be missing or corrupted, please check error log.\n')
text_widget.write(base_text + 'Moving on...\n')
else:
text_widget.write(base_text + 'Failed to save output(s) as Mp3(s).\n')
text_widget.write(base_text + 'Please check error log.\n')
text_widget.write(base_text + 'Moving on...\n')
try:
with open('errorlog.txt', 'w') as f:
f.write(f'Last Error Received:\n\n' +
f'Error Received while attempting to save file as mp3 "{os.path.basename(music_file)}":\n' +
f'Process Method: VR Architecture\n\n' +
f'FFmpeg might be missing or corrupted.\n\n' +
f'If this error persists, please contact the developers.\n\n' +
f'Raw error details:\n\n' +
errmessage + f'\nError Time Stamp: [{datetime.now().strftime("%Y-%m-%d %H:%M:%S")}]\n')
except:
pass
if data['saveFormat'] == 'Flac':
try:
if VModel in model_name:
if data['inst_only'] == True:
pass
else:
musfile = pydub.AudioSegment.from_wav(instrumental_path)
musfile.export(instrumental_path_flac, format="flac")
if file_exists_v == 'there':
pass
else:
try:
os.remove(instrumental_path)
except:
pass
if data['voc_only'] == True:
pass
else:
musfile = pydub.AudioSegment.from_wav(vocal_path)
musfile.export(vocal_path_flac, format="flac")
if file_exists_i == 'there':
pass
else:
try:
os.remove(vocal_path)
except:
pass
else:
if data['inst_only'] == True:
pass
else:
musfile = pydub.AudioSegment.from_wav(vocal_path)
musfile.export(vocal_path_flac, format="flac")
if file_exists_v == 'there':
pass
else:
try:
os.remove(vocal_path)
except:
pass
if data['voc_only'] == True:
pass
else:
musfile = pydub.AudioSegment.from_wav(instrumental_path)
musfile.export(instrumental_path_flac, format="flac")
if file_exists_i == 'there':
pass
else:
try:
os.remove(instrumental_path)
except:
pass
except Exception as e:
traceback_text = ''.join(traceback.format_tb(e.__traceback__))
errmessage = f'Traceback Error: "{traceback_text}"\n{type(e).__name__}: "{e}"\n'
if "ffmpeg" in errmessage:
text_widget.write(base_text + 'Failed to save output(s) as Flac(s).\n')
text_widget.write(base_text + 'FFmpeg might be missing or corrupted, please check error log.\n')
text_widget.write(base_text + 'Moving on...\n')
else:
text_widget.write(base_text + 'Failed to save output(s) as Flac(s).\n')
text_widget.write(base_text + 'Please check error log.\n')
text_widget.write(base_text + 'Moving on...\n')
try:
with open('errorlog.txt', 'w') as f:
f.write(f'Last Error Received:\n\n' +
f'Error Received while attempting to save file as flac "{os.path.basename(music_file)}":\n' +
f'Process Method: VR Architecture\n\n' +
f'FFmpeg might be missing or corrupted.\n\n' +
f'If this error persists, please contact the developers.\n\n' +
f'Raw error details:\n\n' +
errmessage + f'\nError Time Stamp: [{datetime.now().strftime("%Y-%m-%d %H:%M:%S")}]\n')
except:
pass
data.update(kwargs)
# Update default settings
global default_window_size
global default_agg
global normalization_set
default_window_size = data['window_size']
default_agg = data['agg']
stime = time.perf_counter()
progress_var.set(0)
text_widget.clear()
button_widget.configure(state=tk.DISABLED) # Disable Button
overlap_set = float(data['overlap'])
shift_set = int(data['shifts'])
demucs_model_set = data['demucsmodel_sel_VR']
split_mode = data['split_mode']
if data['wavtype'] == '32-bit Float':
wav_type_set = 'FLOAT'
elif data['wavtype'] == '64-bit Float':
wav_type_set = 'DOUBLE'
else:
wav_type_set = data['wavtype']
flac_type_set = data['flactype']
mp3_bit_set = data['mp3bit']
if data['normalize'] == True:
normalization_set = spec_utils.normalize
print('normalization on')
else:
normalization_set = spec_utils.nonormalize
print('normalization off')
vocal_remover = VocalRemover(data, text_widget)
modelFolderName = determineModelFolderName()
timestampnum = round(datetime.utcnow().timestamp())
randomnum = randrange(100000, 1000000)
# Separation Preperation
try: #Load File(s)
for file_num, music_file in enumerate(data['input_paths'], start=1):
# Determine File Name
m=music_file
if data['settest']:
try:
base_name = f'{data["export_path"]}/{str(timestampnum)}_{file_num}_{os.path.splitext(os.path.basename(music_file))[0]}'
except:
base_name = f'{data["export_path"]}/{str(randomnum)}_{file_num}_{os.path.splitext(os.path.basename(music_file))[0]}'
else:
base_name = f'{data["export_path"]}/{file_num}_{os.path.splitext(os.path.basename(music_file))[0]}'
model_name = os.path.basename(data[f'{data["useModel"]}Model'])
model = vocal_remover.models[data['useModel']]
device = vocal_remover.devices[data['useModel']]
# -Get text and update progress-
base_text = get_baseText(total_files=len(data['input_paths']),
file_num=file_num)
progress_kwargs = {'progress_var': progress_var,
'total_files': len(data['input_paths']),
'file_num': file_num}
update_progress(**progress_kwargs,
step=0)
try:
total, used, free = shutil.disk_usage("/")
total_space = int(total/1.074e+9)
used_space = int(used/1.074e+9)
free_space = int(free/1.074e+9)
if int(free/1.074e+9) <= int(2):
text_widget.write('Error: Not enough storage on main drive to continue. Your main drive must have \nat least 3 GB\'s of storage in order for this application function properly. \n\nPlease ensure your main drive has at least 3 GB\'s of storage and try again.\n\n')
text_widget.write('Detected Total Space: ' + str(total_space) + ' GB' + '\n')
text_widget.write('Detected Used Space: ' + str(used_space) + ' GB' + '\n')
text_widget.write('Detected Free Space: ' + str(free_space) + ' GB' + '\n')
progress_var.set(0)
button_widget.configure(state=tk.NORMAL) # Enable Button
return
if int(free/1.074e+9) in [3, 4, 5, 6, 7, 8]:
text_widget.write('Warning: Your main drive is running low on storage. Your main drive must have \nat least 3 GB\'s of storage in order for this application function properly.\n\n')
text_widget.write('Detected Total Space: ' + str(total_space) + ' GB' + '\n')
text_widget.write('Detected Used Space: ' + str(used_space) + ' GB' + '\n')
text_widget.write('Detected Free Space: ' + str(free_space) + ' GB' + '\n\n')
except:
pass
if data['wavtype'] == '64-bit Float':
if data['saveFormat'] == 'Flac':
text_widget.write('Please select \"WAV\" as your save format to use 64-bit Float.\n')
text_widget.write(f'Time Elapsed: {time.strftime("%H:%M:%S", time.gmtime(int(time.perf_counter() - stime)))}')
progress_var.set(0)
button_widget.configure(state=tk.NORMAL) # Enable Button
return
if data['wavtype'] == '64-bit Float':
if data['saveFormat'] == 'Mp3':
text_widget.write('Please select \"WAV\" as your save format to use 64-bit Float.\n')
text_widget.write(f'Time Elapsed: {time.strftime("%H:%M:%S", time.gmtime(int(time.perf_counter() - stime)))}')
progress_var.set(0)
button_widget.configure(state=tk.NORMAL) # Enable Button
return
#Load Model
text_widget.write(base_text + 'Loading model...')
model_size = math.ceil(os.stat(data['instrumentalModel']).st_size / 1024)
nn_architecture = '{}KB'.format(min(nn_arch_sizes, key=lambda x:abs(x-model_size)))
nets = importlib.import_module('lib_v5.nets' + f'_{nn_architecture}'.replace('_{}KB'.format(nn_arch_sizes[0]), ''), package=None)
aggresive_set = float(data['agg']/100)
ModelName=(data['instrumentalModel'])
#Package Models
text_widget.write('Done!\n')
if data['ModelParams'] == 'Auto':
model_hash = hashlib.md5(open(ModelName,'rb').read()).hexdigest()
model_params = []
model_params = lib_v5.filelist.provide_model_param_hash(model_hash)
print(model_params)
if model_params[0] == 'Not Found Using Hash':
model_params = []
model_params = lib_v5.filelist.provide_model_param_name(ModelName)
if model_params[0] == 'Not Found Using Name':
text_widget.write(base_text + f'Unable to set model parameters automatically with the selected model.\n')
confirm = tk.messagebox.askyesno(title='Unrecognized Model Detected',
message=f'\nThe application could not automatically set the model param for the selected model.\n\n' +
f'Would you like to select the model param file for this model?\n\n')
if confirm:
model_param_selection = filedialog.askopenfilename(initialdir='lib_v5/modelparams',
title=f'Select Model Param',
filetypes=[("Model Param", "*.json")])
model_param_file_path = str(model_param_selection)
model_param_file = os.path.splitext(os.path.basename(model_param_file_path))[0] + '.json'
model_params = [model_param_file_path, model_param_file]
with open(f"lib_v5/filelists/model_cache/vr_param_cache/{model_hash}.txt", 'w') as f:
f.write(model_param_file)
if model_params[0] == '':
text_widget.write("\n" + base_text + f'Separation failed for the following audio file:\n')
text_widget.write(base_text + f'"{os.path.basename(music_file)}"\n')
text_widget.write(f'\nError Received:\n\n')
text_widget.write(f'Model parameters are missing.\n\n')
text_widget.write(f'Please check the following:\n')
text_widget.write(f'1. Make sure the model is still present.\n')
text_widget.write(f'2. If you are running a model that was not originally included in this package, \nplease append the modelparam name to the model name.\n')
text_widget.write(f' - Example if using \"4band_v2.json\" modelparam: \"model_4band_v2.pth\"\n\n')
text_widget.write(f'Please address this and try again.\n\n')
text_widget.write(f'Time Elapsed: {time.strftime("%H:%M:%S", time.gmtime(int(time.perf_counter() - stime)))}')
torch.cuda.empty_cache()
progress_var.set(0)
button_widget.configure(state=tk.NORMAL) # Enable Button
return
else:
pass
else:
text_widget.write(base_text + f'Model param not selected.\n')
text_widget.write("\n" + base_text + f'Separation failed for the following audio file:\n')
text_widget.write(base_text + f'"{os.path.basename(music_file)}"\n')
text_widget.write(f'\nError Received:\n\n')
text_widget.write(f'Model parameters are missing.\n\n')
text_widget.write(f'Please check the following:\n')
text_widget.write(f'1. Make sure the model is still present.\n')
text_widget.write(f'2. If you are running a model that was not originally included in this package, \nplease append the modelparam name to the model name.\n')
text_widget.write(f' - Example if using \"4band_v2.json\" modelparam: \"model_4band_v2.pth\"\n\n')
text_widget.write(f'Please address this and try again.\n\n')
text_widget.write(f'Time Elapsed: {time.strftime("%H:%M:%S", time.gmtime(int(time.perf_counter() - stime)))}')
torch.cuda.empty_cache()
progress_var.set(0)
button_widget.configure(state=tk.NORMAL) # Enable Button
return
else:
param = data['ModelParams']
model_param_file_path = f'lib_v5/modelparams/{param}'
model_params = [model_param_file_path, param]
text_widget.write(base_text + 'Loading assigned model parameters ' + '\"' + model_params[1] + '\"... ')
mp = ModelParameters(model_params[0])
text_widget.write('Done!\n')
# -Instrumental-
if os.path.isfile(data['instrumentalModel']):
device = torch.device('cpu')
model = nets.CascadedASPPNet(mp.param['bins'] * 2)
model.load_state_dict(torch.load(data['instrumentalModel'],
map_location=device))
if torch.cuda.is_available() and data['gpu'] >= 0:
device = torch.device('cuda:{}'.format(data['gpu']))
model.to(device)
vocal_remover.models['instrumental'] = model
vocal_remover.devices['instrumental'] = device
model_name = os.path.basename(data[f'{data["useModel"]}Model'])
# -Go through the different steps of Separation-
# Wave source
text_widget.write(base_text + 'Loading audio source...')
X_wave, y_wave, X_spec_s, y_spec_s = {}, {}, {}, {}
bands_n = len(mp.param['band'])
for d in range(bands_n, 0, -1):
bp = mp.param['band'][d]
if d == bands_n: # high-end band
X_wave[d], _ = librosa.load(
music_file, bp['sr'], False, dtype=np.float32, res_type=bp['res_type'])
if X_wave[d].ndim == 1:
X_wave[d] = np.asarray([X_wave[d], X_wave[d]])
else: # lower bands
X_wave[d] = librosa.resample(X_wave[d+1], mp.param['band'][d+1]['sr'], bp['sr'], res_type=bp['res_type'])
# Stft of wave source
X_spec_s[d] = spec_utils.wave_to_spectrogram_mt(X_wave[d], bp['hl'], bp['n_fft'], mp.param['mid_side'],
mp.param['mid_side_b2'], mp.param['reverse'])
if d == bands_n and data['high_end_process'] != 'none':
input_high_end_h = (bp['n_fft']//2 - bp['crop_stop']) + (mp.param['pre_filter_stop'] - mp.param['pre_filter_start'])
input_high_end = X_spec_s[d][:, bp['n_fft']//2-input_high_end_h:bp['n_fft']//2, :]
text_widget.write('Done!\n')
update_progress(**progress_kwargs,
step=0.1)
text_widget.write(base_text + 'Loading the stft of audio source...')
text_widget.write(' Done!\n')
text_widget.write(base_text + "Please Wait...\n")
X_spec_m = spec_utils.combine_spectrograms(X_spec_s, mp)
del X_wave, X_spec_s
def inference(X_spec, device, model, aggressiveness):
def _execute(X_mag_pad, roi_size, n_window, device, model, aggressiveness):
model.eval()
with torch.no_grad():
preds = []
iterations = [n_window]
total_iterations = sum(iterations)
text_widget.write(base_text + "Processing "f"{total_iterations} Slices... ")
for i in tqdm(range(n_window)):
update_progress(**progress_kwargs,
step=(0.1 + (0.8/n_window * i)))
start = i * roi_size
X_mag_window = X_mag_pad[None, :, :, start:start + data['window_size']]
X_mag_window = torch.from_numpy(X_mag_window).to(device)
pred = model.predict(X_mag_window, aggressiveness)
pred = pred.detach().cpu().numpy()
preds.append(pred[0])
pred = np.concatenate(preds, axis=2)
text_widget.write('Done!\n')
return pred
def preprocess(X_spec):
X_mag = np.abs(X_spec)
X_phase = np.angle(X_spec)
return X_mag, X_phase
X_mag, X_phase = preprocess(X_spec)
coef = X_mag.max()
X_mag_pre = X_mag / coef
n_frame = X_mag_pre.shape[2]
pad_l, pad_r, roi_size = dataset.make_padding(n_frame,
data['window_size'], model.offset)
n_window = int(np.ceil(n_frame / roi_size))
X_mag_pad = np.pad(
X_mag_pre, ((0, 0), (0, 0), (pad_l, pad_r)), mode='constant')
pred = _execute(X_mag_pad, roi_size, n_window,
device, model, aggressiveness)
pred = pred[:, :, :n_frame]
if data['tta']:
pad_l += roi_size // 2
pad_r += roi_size // 2
n_window += 1
X_mag_pad = np.pad(
X_mag_pre, ((0, 0), (0, 0), (pad_l, pad_r)), mode='constant')
pred_tta = _execute(X_mag_pad, roi_size, n_window,
device, model, aggressiveness)
pred_tta = pred_tta[:, :, roi_size // 2:]
pred_tta = pred_tta[:, :, :n_frame]
return (pred + pred_tta) * 0.5 * coef, X_mag, np.exp(1.j * X_phase)
else:
return pred * coef, X_mag, np.exp(1.j * X_phase)
aggressiveness = {'value': aggresive_set, 'split_bin': mp.param['band'][1]['crop_stop']}
if data['tta']:
text_widget.write(base_text + "Running Inferences (TTA)...\n")
else:
text_widget.write(base_text + "Running Inference...\n")
pred, X_mag, X_phase = inference(X_spec_m,
device,
model, aggressiveness)
update_progress(**progress_kwargs,
step=0.9)
# Postprocess
if data['postprocess']:
try:
text_widget.write(base_text + 'Post processing...')
pred_inv = np.clip(X_mag - pred, 0, np.inf)
pred = spec_utils.mask_silence(pred, pred_inv)
text_widget.write(' Done!\n')
except Exception as e:
text_widget.write('\n' + base_text + 'Post process failed, check error log.\n')
text_widget.write(base_text + 'Moving on...\n')
traceback_text = ''.join(traceback.format_tb(e.__traceback__))
errmessage = f'Traceback Error: "{traceback_text}"\n{type(e).__name__}: "{e}"\n'
try:
with open('errorlog.txt', 'w') as f:
f.write(f'Last Error Received:\n\n' +
f'Error Received while attempting to run Post Processing on "{os.path.basename(music_file)}":\n' +
f'Process Method: VR Architecture\n\n' +
f'If this error persists, please contact the developers.\n\n' +
f'Raw error details:\n\n' +
errmessage + f'\nError Time Stamp: [{datetime.now().strftime("%Y-%m-%d %H:%M:%S")}]\n')
except:
pass
update_progress(**progress_kwargs,
step=0.95)
# Inverse stft
y_spec_m = pred * X_phase
v_spec_m = X_spec_m - y_spec_m
def demix_demucs(mix):
#print('shift_set ', shift_set)
text_widget.write(base_text + "Running Demucs Inference...\n")
text_widget.write(base_text + "Processing... ")
print(' Running Demucs Inference...')
mix = torch.tensor(mix, dtype=torch.float32)
ref = mix.mean(0)
mix = (mix - ref.mean()) / ref.std()
with torch.no_grad():
sources = apply_model(demucs, mix[None], split=split_mode, device=device, overlap=overlap_set, shifts=shift_set, progress=False)[0]
text_widget.write('Done!\n')
sources = (sources * ref.std() + ref.mean()).cpu().numpy()
sources[[0,1]] = sources[[1,0]]
return sources
def demucs_prediction(m):
global demucs_sources
mix, samplerate = librosa.load(m, mono=False, sr=44100)
if mix.ndim == 1:
mix = np.asfortranarray([mix,mix])
mix = mix.T
demucs_sources = demix_demucs(mix.T)
if data['demucsmodelVR']:
demucs = HDemucs(sources=["other", "vocals"])
text_widget.write(base_text + 'Loading Demucs model... ')
update_progress(**progress_kwargs,
step=0.95)
path_d = Path('models/Demucs_Models/v3_repo')
#print('What Demucs model was chosen? ', demucs_model_set)
demucs = _gm(name=demucs_model_set, repo=path_d)
text_widget.write('Done!\n')
#print('segment: ', data['segment'])
if data['segment'] == 'None':
segment = None
if isinstance(demucs, BagOfModels):
if segment is not None:
for sub in demucs.models:
sub.segment = segment
else:
if segment is not None:
sub.segment = segment
else:
try:
segment = int(data['segment'])
if isinstance(demucs, BagOfModels):
if segment is not None:
for sub in demucs.models:
sub.segment = segment
else:
if segment is not None:
sub.segment = segment
text_widget.write(base_text + "Segments set to "f"{segment}.\n")
except:
segment = None
if isinstance(demucs, BagOfModels):
if segment is not None:
for sub in demucs.models:
sub.segment = segment
else:
if segment is not None:
sub.segment = segment
#print('segment port-process: ', segment)
demucs.cpu()
demucs.eval()
demucs_prediction(m)
if data['voc_only'] and not data['inst_only']:
pass
else:
text_widget.write(base_text + 'Saving Instrumental... ')
if data['high_end_process'].startswith('mirroring'):
input_high_end_ = spec_utils.mirroring(data['high_end_process'], y_spec_m, input_high_end, mp)
if data['demucsmodelVR']:
wav_instrument = spec_utils.cmb_spectrogram_to_wave_d(y_spec_m, mp, input_high_end_h, input_high_end_, demucs=True)
demucs_inst = demucs_sources[0]
sources = [wav_instrument,demucs_inst]
spec = [stft(sources[0],2048,1024),stft(sources[1],2048,1024)]
ln = min([spec[0].shape[2], spec[1].shape[2]])
spec[0] = spec[0][:,:,:ln]
spec[1] = spec[1][:,:,:ln]
v_spec_c = np.where(np.abs(spec[1]) <= np.abs(spec[0]), spec[1], spec[0])
wav_instrument = istft(v_spec_c,1024)
else:
wav_instrument = spec_utils.cmb_spectrogram_to_wave_d(y_spec_m, mp, input_high_end_h, input_high_end_, demucs=False)
if data['voc_only'] and not data['inst_only']:
pass
else:
text_widget.write('Done!\n')
else:
wav_instrument = spec_utils.cmb_spectrogram_to_wave_d(y_spec_m, mp)
if data['voc_only'] and not data['inst_only']:
pass
else:
text_widget.write('Done!\n')
if data['inst_only'] and not data['voc_only']:
pass
else:
text_widget.write(base_text + 'Saving Vocals... ')
if data['high_end_process'].startswith('mirroring'):
input_high_end_ = spec_utils.mirroring(data['high_end_process'], v_spec_m, input_high_end, mp)
if data['demucsmodelVR']:
wav_vocals = spec_utils.cmb_spectrogram_to_wave_d(v_spec_m, mp, input_high_end_h, input_high_end_, demucs=True)
demucs_voc = demucs_sources[1]
sources = [wav_vocals,demucs_voc]
spec = [stft(sources[0],2048,1024),stft(sources[1],2048,1024)]
ln = min([spec[0].shape[2], spec[1].shape[2]])
spec[0] = spec[0][:,:,:ln]
spec[1] = spec[1][:,:,:ln]
v_spec_c = np.where(np.abs(spec[1]) >= np.abs(spec[0]), spec[1], spec[0])
wav_vocals = istft(v_spec_c,1024)
else:
wav_vocals = spec_utils.cmb_spectrogram_to_wave_d(v_spec_m, mp, input_high_end_h, input_high_end_, demucs=False)
if data['inst_only'] and not data['voc_only']:
pass
else:
text_widget.write('Done!\n')
else:
wav_vocals = spec_utils.cmb_spectrogram_to_wave_d(v_spec_m, mp, demucs=False)
if data['inst_only'] and not data['voc_only']:
pass
else:
text_widget.write('Done!\n')
update_progress(**progress_kwargs,
step=1)
# Save output music files
save_files(wav_instrument, wav_vocals)
update_progress(**progress_kwargs,
step=1)
# Save output image
if data['output_image']:
with open('{}_Instruments.jpg'.format(base_name), mode='wb') as f:
image = spec_utils.spectrogram_to_image(y_spec_m)
_, bin_image = cv2.imencode('.jpg', image)
bin_image.tofile(f)
with open('{}_Vocals.jpg'.format(base_name), mode='wb') as f:
image = spec_utils.spectrogram_to_image(v_spec_m)
_, bin_image = cv2.imencode('.jpg', image)
bin_image.tofile(f)
text_widget.write(base_text + 'Completed Separation!\n\n')
except Exception as e:
traceback_text = ''.join(traceback.format_tb(e.__traceback__))
message = f'Traceback Error: "{traceback_text}"\n{type(e).__name__}: "{e}"\n'
if runtimeerr in message:
text_widget.write("\n" + base_text + f'Separation failed for the following audio file:\n')
text_widget.write(base_text + f'"{os.path.basename(music_file)}"\n')
text_widget.write(f'\nError Received:\n\n')
text_widget.write(f'Your PC cannot process this audio file with the chunk size selected.\nPlease lower the chunk size and try again.\n\n')
text_widget.write(f'If this error persists, please contact the developers.\n\n')
text_widget.write(f'Time Elapsed: {time.strftime("%H:%M:%S", time.gmtime(int(time.perf_counter() - stime)))}')
try:
with open('errorlog.txt', 'w') as f:
f.write(f'Last Error Received:\n\n' +
f'Error Received while processing "{os.path.basename(music_file)}":\n' +
f'Process Method: VR Architecture\n\n' +
f'Your PC cannot process this audio file with the chunk size selected.\nPlease lower the chunk size and try again.\n\n' +
f'If this error persists, please contact the developers.\n\n' +
f'Raw error details:\n\n' +
message + f'\nError Time Stamp: [{datetime.now().strftime("%Y-%m-%d %H:%M:%S")}]\n')
except:
pass
torch.cuda.empty_cache()
progress_var.set(0)
button_widget.configure(state=tk.NORMAL) # Enable Button
return
if cuda_err in message:
text_widget.write("\n" + base_text + f'Separation failed for the following audio file:\n')
text_widget.write(base_text + f'"{os.path.basename(music_file)}"\n')
text_widget.write(f'\nError Received:\n\n')
text_widget.write(f'The application was unable to allocate enough GPU memory to use this model.\n')
text_widget.write(f'Please close any GPU intensive applications and try again.\n')
text_widget.write(f'If the error persists, your GPU might not be supported.\n\n')
text_widget.write(f'Time Elapsed: {time.strftime("%H:%M:%S", time.gmtime(int(time.perf_counter() - stime)))}')
try:
with open('errorlog.txt', 'w') as f:
f.write(f'Last Error Received:\n\n' +
f'Error Received while processing "{os.path.basename(music_file)}":\n' +
f'Process Method: VR Architecture\n\n' +
f'The application was unable to allocate enough GPU memory to use this model.\n' +
f'Please close any GPU intensive applications and try again.\n' +
f'If the error persists, your GPU might not be supported.\n\n' +
f'Raw error details:\n\n' +
message + f'\nError Time Stamp [{datetime.now().strftime("%Y-%m-%d %H:%M:%S")}]\n')
except:
pass
torch.cuda.empty_cache()
progress_var.set(0)
button_widget.configure(state=tk.NORMAL) # Enable Button
return
if mod_err in message:
text_widget.write("\n" + base_text + f'Separation failed for the following audio file:\n')
text_widget.write(base_text + f'"{os.path.basename(music_file)}"\n')
text_widget.write(f'\nError Received:\n\n')
text_widget.write(f'Application files(s) are missing.\n')
text_widget.write("\n" + f'{type(e).__name__} - "{e}"' + "\n\n")
text_widget.write(f'Please check for missing files/scripts in the app directory and try again.\n')
text_widget.write(f'If the error persists, please reinstall application or contact the developers.\n\n')
text_widget.write(f'Time Elapsed: {time.strftime("%H:%M:%S", time.gmtime(int(time.perf_counter() - stime)))}')
try:
with open('errorlog.txt', 'w') as f:
f.write(f'Last Error Received:\n\n' +
f'Error Received while processing "{os.path.basename(music_file)}":\n' +
f'Process Method: VR Architecture\n\n' +
f'Application files(s) are missing.\n' +
f'Please check for missing files/scripts in the app directory and try again.\n' +
f'If the error persists, please reinstall application or contact the developers.\n\n' +
message + f'\nError Time Stamp [{datetime.now().strftime("%Y-%m-%d %H:%M:%S")}]\n')
except:
pass
torch.cuda.empty_cache()
progress_var.set(0)
button_widget.configure(state=tk.NORMAL) # Enable Button
return
if file_err in message:
text_widget.write("\n" + base_text + f'Separation failed for the following audio file:\n')
text_widget.write(base_text + f'"{os.path.basename(music_file)}"\n')
text_widget.write(f'\nError Received:\n\n')
text_widget.write(f'Missing file error raised.\n')
text_widget.write("\n" + f'{type(e).__name__} - "{e}"' + "\n\n")
text_widget.write("\n" + f'Please address the error and try again.' + "\n")
text_widget.write(f'If this error persists, please contact the developers.\n\n')
text_widget.write(f'Time Elapsed: {time.strftime("%H:%M:%S", time.gmtime(int(time.perf_counter() - stime)))}')
torch.cuda.empty_cache()
try:
with open('errorlog.txt', 'w') as f:
f.write(f'Last Error Received:\n\n' +
f'Error Received while processing "{os.path.basename(music_file)}":\n' +
f'Process Method: VR Architecture\n\n' +
f'Missing file error raised.\n' +
"\n" + f'Please address the error and try again.' + "\n" +