File size: 43,650 Bytes
ffead1e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
# coding=utf-8
# Copyright 2023 The HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
""" Conversion script for the AudioLDM checkpoints."""

import argparse
import re

import torch
from transformers import (
    AutoTokenizer,
    ClapTextConfig,
    ClapTextModelWithProjection,
    SpeechT5HifiGan,
    SpeechT5HifiGanConfig,
)

from diffusers import (
    AudioLDMPipeline,
    AutoencoderKL,
    DDIMScheduler,
    DPMSolverMultistepScheduler,
    EulerAncestralDiscreteScheduler,
    EulerDiscreteScheduler,
    HeunDiscreteScheduler,
    LMSDiscreteScheduler,
    PNDMScheduler,
    UNet2DConditionModel,
)
from diffusers.utils import is_omegaconf_available, is_safetensors_available
from diffusers.utils.import_utils import BACKENDS_MAPPING


# Copied from diffusers.pipelines.stable_diffusion.convert_from_ckpt.shave_segments
def shave_segments(path, n_shave_prefix_segments=1):
    """
    Removes segments. Positive values shave the first segments, negative shave the last segments.
    """
    if n_shave_prefix_segments >= 0:
        return ".".join(path.split(".")[n_shave_prefix_segments:])
    else:
        return ".".join(path.split(".")[:n_shave_prefix_segments])


# Copied from diffusers.pipelines.stable_diffusion.convert_from_ckpt.renew_resnet_paths
def renew_resnet_paths(old_list, n_shave_prefix_segments=0):
    """
    Updates paths inside resnets to the new naming scheme (local renaming)
    """
    mapping = []
    for old_item in old_list:
        new_item = old_item.replace("in_layers.0", "norm1")
        new_item = new_item.replace("in_layers.2", "conv1")

        new_item = new_item.replace("out_layers.0", "norm2")
        new_item = new_item.replace("out_layers.3", "conv2")

        new_item = new_item.replace("emb_layers.1", "time_emb_proj")
        new_item = new_item.replace("skip_connection", "conv_shortcut")

        new_item = shave_segments(new_item, n_shave_prefix_segments=n_shave_prefix_segments)

        mapping.append({"old": old_item, "new": new_item})

    return mapping


# Copied from diffusers.pipelines.stable_diffusion.convert_from_ckpt.renew_vae_resnet_paths
def renew_vae_resnet_paths(old_list, n_shave_prefix_segments=0):
    """
    Updates paths inside resnets to the new naming scheme (local renaming)
    """
    mapping = []
    for old_item in old_list:
        new_item = old_item

        new_item = new_item.replace("nin_shortcut", "conv_shortcut")
        new_item = shave_segments(new_item, n_shave_prefix_segments=n_shave_prefix_segments)

        mapping.append({"old": old_item, "new": new_item})

    return mapping


# Copied from diffusers.pipelines.stable_diffusion.convert_from_ckpt.renew_attention_paths
def renew_attention_paths(old_list):
    """
    Updates paths inside attentions to the new naming scheme (local renaming)
    """
    mapping = []
    for old_item in old_list:
        new_item = old_item

        #         new_item = new_item.replace('norm.weight', 'group_norm.weight')
        #         new_item = new_item.replace('norm.bias', 'group_norm.bias')

        #         new_item = new_item.replace('proj_out.weight', 'proj_attn.weight')
        #         new_item = new_item.replace('proj_out.bias', 'proj_attn.bias')

        #         new_item = shave_segments(new_item, n_shave_prefix_segments=n_shave_prefix_segments)

        mapping.append({"old": old_item, "new": new_item})

    return mapping


# Copied from diffusers.pipelines.stable_diffusion.convert_from_ckpt.renew_vae_attention_paths
def renew_vae_attention_paths(old_list, n_shave_prefix_segments=0):
    """
    Updates paths inside attentions to the new naming scheme (local renaming)
    """
    mapping = []
    for old_item in old_list:
        new_item = old_item

        new_item = new_item.replace("norm.weight", "group_norm.weight")
        new_item = new_item.replace("norm.bias", "group_norm.bias")

        new_item = new_item.replace("q.weight", "query.weight")
        new_item = new_item.replace("q.bias", "query.bias")

        new_item = new_item.replace("k.weight", "key.weight")
        new_item = new_item.replace("k.bias", "key.bias")

        new_item = new_item.replace("v.weight", "value.weight")
        new_item = new_item.replace("v.bias", "value.bias")

        new_item = new_item.replace("proj_out.weight", "proj_attn.weight")
        new_item = new_item.replace("proj_out.bias", "proj_attn.bias")

        new_item = shave_segments(new_item, n_shave_prefix_segments=n_shave_prefix_segments)

        mapping.append({"old": old_item, "new": new_item})

    return mapping


# Copied from diffusers.pipelines.stable_diffusion.convert_from_ckpt.assign_to_checkpoint
def assign_to_checkpoint(
    paths, checkpoint, old_checkpoint, attention_paths_to_split=None, additional_replacements=None, config=None
):
    """
    This does the final conversion step: take locally converted weights and apply a global renaming to them. It splits
    attention layers, and takes into account additional replacements that may arise.

    Assigns the weights to the new checkpoint.
    """
    assert isinstance(paths, list), "Paths should be a list of dicts containing 'old' and 'new' keys."

    # Splits the attention layers into three variables.
    if attention_paths_to_split is not None:
        for path, path_map in attention_paths_to_split.items():
            old_tensor = old_checkpoint[path]
            channels = old_tensor.shape[0] // 3

            target_shape = (-1, channels) if len(old_tensor.shape) == 3 else (-1)

            num_heads = old_tensor.shape[0] // config["num_head_channels"] // 3

            old_tensor = old_tensor.reshape((num_heads, 3 * channels // num_heads) + old_tensor.shape[1:])
            query, key, value = old_tensor.split(channels // num_heads, dim=1)

            checkpoint[path_map["query"]] = query.reshape(target_shape)
            checkpoint[path_map["key"]] = key.reshape(target_shape)
            checkpoint[path_map["value"]] = value.reshape(target_shape)

    for path in paths:
        new_path = path["new"]

        # These have already been assigned
        if attention_paths_to_split is not None and new_path in attention_paths_to_split:
            continue

        # Global renaming happens here
        new_path = new_path.replace("middle_block.0", "mid_block.resnets.0")
        new_path = new_path.replace("middle_block.1", "mid_block.attentions.0")
        new_path = new_path.replace("middle_block.2", "mid_block.resnets.1")

        if additional_replacements is not None:
            for replacement in additional_replacements:
                new_path = new_path.replace(replacement["old"], replacement["new"])

        # proj_attn.weight has to be converted from conv 1D to linear
        if "proj_attn.weight" in new_path:
            checkpoint[new_path] = old_checkpoint[path["old"]][:, :, 0]
        else:
            checkpoint[new_path] = old_checkpoint[path["old"]]


# Copied from diffusers.pipelines.stable_diffusion.convert_from_ckpt.conv_attn_to_linear
def conv_attn_to_linear(checkpoint):
    keys = list(checkpoint.keys())
    attn_keys = ["query.weight", "key.weight", "value.weight"]
    for key in keys:
        if ".".join(key.split(".")[-2:]) in attn_keys:
            if checkpoint[key].ndim > 2:
                checkpoint[key] = checkpoint[key][:, :, 0, 0]
        elif "proj_attn.weight" in key:
            if checkpoint[key].ndim > 2:
                checkpoint[key] = checkpoint[key][:, :, 0]


def create_unet_diffusers_config(original_config, image_size: int):
    """
    Creates a UNet config for diffusers based on the config of the original AudioLDM model.
    """
    unet_params = original_config.model.params.unet_config.params
    vae_params = original_config.model.params.first_stage_config.params.ddconfig

    block_out_channels = [unet_params.model_channels * mult for mult in unet_params.channel_mult]

    down_block_types = []
    resolution = 1
    for i in range(len(block_out_channels)):
        block_type = "CrossAttnDownBlock2D" if resolution in unet_params.attention_resolutions else "DownBlock2D"
        down_block_types.append(block_type)
        if i != len(block_out_channels) - 1:
            resolution *= 2

    up_block_types = []
    for i in range(len(block_out_channels)):
        block_type = "CrossAttnUpBlock2D" if resolution in unet_params.attention_resolutions else "UpBlock2D"
        up_block_types.append(block_type)
        resolution //= 2

    vae_scale_factor = 2 ** (len(vae_params.ch_mult) - 1)

    cross_attention_dim = (
        unet_params.cross_attention_dim if "cross_attention_dim" in unet_params else block_out_channels
    )

    class_embed_type = "simple_projection" if "extra_film_condition_dim" in unet_params else None
    projection_class_embeddings_input_dim = (
        unet_params.extra_film_condition_dim if "extra_film_condition_dim" in unet_params else None
    )
    class_embeddings_concat = unet_params.extra_film_use_concat if "extra_film_use_concat" in unet_params else None

    config = {
        "sample_size": image_size // vae_scale_factor,
        "in_channels": unet_params.in_channels,
        "out_channels": unet_params.out_channels,
        "down_block_types": tuple(down_block_types),
        "up_block_types": tuple(up_block_types),
        "block_out_channels": tuple(block_out_channels),
        "layers_per_block": unet_params.num_res_blocks,
        "cross_attention_dim": cross_attention_dim,
        "class_embed_type": class_embed_type,
        "projection_class_embeddings_input_dim": projection_class_embeddings_input_dim,
        "class_embeddings_concat": class_embeddings_concat,
    }

    return config


# Adapted from diffusers.pipelines.stable_diffusion.convert_from_ckpt.create_vae_diffusers_config
def create_vae_diffusers_config(original_config, checkpoint, image_size: int):
    """
    Creates a VAE config for diffusers based on the config of the original AudioLDM model. Compared to the original
    Stable Diffusion conversion, this function passes a *learnt* VAE scaling factor to the diffusers VAE.
    """
    vae_params = original_config.model.params.first_stage_config.params.ddconfig
    _ = original_config.model.params.first_stage_config.params.embed_dim

    block_out_channels = [vae_params.ch * mult for mult in vae_params.ch_mult]
    down_block_types = ["DownEncoderBlock2D"] * len(block_out_channels)
    up_block_types = ["UpDecoderBlock2D"] * len(block_out_channels)

    scaling_factor = checkpoint["scale_factor"] if "scale_by_std" in original_config.model.params else 0.18215

    config = {
        "sample_size": image_size,
        "in_channels": vae_params.in_channels,
        "out_channels": vae_params.out_ch,
        "down_block_types": tuple(down_block_types),
        "up_block_types": tuple(up_block_types),
        "block_out_channels": tuple(block_out_channels),
        "latent_channels": vae_params.z_channels,
        "layers_per_block": vae_params.num_res_blocks,
        "scaling_factor": float(scaling_factor),
    }
    return config


# Copied from diffusers.pipelines.stable_diffusion.convert_from_ckpt.create_diffusers_schedular
def create_diffusers_schedular(original_config):
    schedular = DDIMScheduler(
        num_train_timesteps=original_config.model.params.timesteps,
        beta_start=original_config.model.params.linear_start,
        beta_end=original_config.model.params.linear_end,
        beta_schedule="scaled_linear",
    )
    return schedular


# Adapted from diffusers.pipelines.stable_diffusion.convert_from_ckpt.convert_ldm_unet_checkpoint
def convert_ldm_unet_checkpoint(checkpoint, config, path=None, extract_ema=False):
    """
    Takes a state dict and a config, and returns a converted checkpoint. Compared to the original Stable Diffusion
    conversion, this function additionally converts the learnt film embedding linear layer.
    """

    # extract state_dict for UNet
    unet_state_dict = {}
    keys = list(checkpoint.keys())

    unet_key = "model.diffusion_model."
    # at least a 100 parameters have to start with `model_ema` in order for the checkpoint to be EMA
    if sum(k.startswith("model_ema") for k in keys) > 100 and extract_ema:
        print(f"Checkpoint {path} has both EMA and non-EMA weights.")
        print(
            "In this conversion only the EMA weights are extracted. If you want to instead extract the non-EMA"
            " weights (useful to continue fine-tuning), please make sure to remove the `--extract_ema` flag."
        )
        for key in keys:
            if key.startswith("model.diffusion_model"):
                flat_ema_key = "model_ema." + "".join(key.split(".")[1:])
                unet_state_dict[key.replace(unet_key, "")] = checkpoint.pop(flat_ema_key)
    else:
        if sum(k.startswith("model_ema") for k in keys) > 100:
            print(
                "In this conversion only the non-EMA weights are extracted. If you want to instead extract the EMA"
                " weights (usually better for inference), please make sure to add the `--extract_ema` flag."
            )

        for key in keys:
            if key.startswith(unet_key):
                unet_state_dict[key.replace(unet_key, "")] = checkpoint.pop(key)

    new_checkpoint = {}

    new_checkpoint["time_embedding.linear_1.weight"] = unet_state_dict["time_embed.0.weight"]
    new_checkpoint["time_embedding.linear_1.bias"] = unet_state_dict["time_embed.0.bias"]
    new_checkpoint["time_embedding.linear_2.weight"] = unet_state_dict["time_embed.2.weight"]
    new_checkpoint["time_embedding.linear_2.bias"] = unet_state_dict["time_embed.2.bias"]

    new_checkpoint["class_embedding.weight"] = unet_state_dict["film_emb.weight"]
    new_checkpoint["class_embedding.bias"] = unet_state_dict["film_emb.bias"]

    new_checkpoint["conv_in.weight"] = unet_state_dict["input_blocks.0.0.weight"]
    new_checkpoint["conv_in.bias"] = unet_state_dict["input_blocks.0.0.bias"]

    new_checkpoint["conv_norm_out.weight"] = unet_state_dict["out.0.weight"]
    new_checkpoint["conv_norm_out.bias"] = unet_state_dict["out.0.bias"]
    new_checkpoint["conv_out.weight"] = unet_state_dict["out.2.weight"]
    new_checkpoint["conv_out.bias"] = unet_state_dict["out.2.bias"]

    # Retrieves the keys for the input blocks only
    num_input_blocks = len({".".join(layer.split(".")[:2]) for layer in unet_state_dict if "input_blocks" in layer})
    input_blocks = {
        layer_id: [key for key in unet_state_dict if f"input_blocks.{layer_id}" in key]
        for layer_id in range(num_input_blocks)
    }

    # Retrieves the keys for the middle blocks only
    num_middle_blocks = len({".".join(layer.split(".")[:2]) for layer in unet_state_dict if "middle_block" in layer})
    middle_blocks = {
        layer_id: [key for key in unet_state_dict if f"middle_block.{layer_id}" in key]
        for layer_id in range(num_middle_blocks)
    }

    # Retrieves the keys for the output blocks only
    num_output_blocks = len({".".join(layer.split(".")[:2]) for layer in unet_state_dict if "output_blocks" in layer})
    output_blocks = {
        layer_id: [key for key in unet_state_dict if f"output_blocks.{layer_id}" in key]
        for layer_id in range(num_output_blocks)
    }

    for i in range(1, num_input_blocks):
        block_id = (i - 1) // (config["layers_per_block"] + 1)
        layer_in_block_id = (i - 1) % (config["layers_per_block"] + 1)

        resnets = [
            key for key in input_blocks[i] if f"input_blocks.{i}.0" in key and f"input_blocks.{i}.0.op" not in key
        ]
        attentions = [key for key in input_blocks[i] if f"input_blocks.{i}.1" in key]

        if f"input_blocks.{i}.0.op.weight" in unet_state_dict:
            new_checkpoint[f"down_blocks.{block_id}.downsamplers.0.conv.weight"] = unet_state_dict.pop(
                f"input_blocks.{i}.0.op.weight"
            )
            new_checkpoint[f"down_blocks.{block_id}.downsamplers.0.conv.bias"] = unet_state_dict.pop(
                f"input_blocks.{i}.0.op.bias"
            )

        paths = renew_resnet_paths(resnets)
        meta_path = {"old": f"input_blocks.{i}.0", "new": f"down_blocks.{block_id}.resnets.{layer_in_block_id}"}
        assign_to_checkpoint(
            paths, new_checkpoint, unet_state_dict, additional_replacements=[meta_path], config=config
        )

        if len(attentions):
            paths = renew_attention_paths(attentions)
            meta_path = {"old": f"input_blocks.{i}.1", "new": f"down_blocks.{block_id}.attentions.{layer_in_block_id}"}
            assign_to_checkpoint(
                paths, new_checkpoint, unet_state_dict, additional_replacements=[meta_path], config=config
            )

    resnet_0 = middle_blocks[0]
    attentions = middle_blocks[1]
    resnet_1 = middle_blocks[2]

    resnet_0_paths = renew_resnet_paths(resnet_0)
    assign_to_checkpoint(resnet_0_paths, new_checkpoint, unet_state_dict, config=config)

    resnet_1_paths = renew_resnet_paths(resnet_1)
    assign_to_checkpoint(resnet_1_paths, new_checkpoint, unet_state_dict, config=config)

    attentions_paths = renew_attention_paths(attentions)
    meta_path = {"old": "middle_block.1", "new": "mid_block.attentions.0"}
    assign_to_checkpoint(
        attentions_paths, new_checkpoint, unet_state_dict, additional_replacements=[meta_path], config=config
    )

    for i in range(num_output_blocks):
        block_id = i // (config["layers_per_block"] + 1)
        layer_in_block_id = i % (config["layers_per_block"] + 1)
        output_block_layers = [shave_segments(name, 2) for name in output_blocks[i]]
        output_block_list = {}

        for layer in output_block_layers:
            layer_id, layer_name = layer.split(".")[0], shave_segments(layer, 1)
            if layer_id in output_block_list:
                output_block_list[layer_id].append(layer_name)
            else:
                output_block_list[layer_id] = [layer_name]

        if len(output_block_list) > 1:
            resnets = [key for key in output_blocks[i] if f"output_blocks.{i}.0" in key]
            attentions = [key for key in output_blocks[i] if f"output_blocks.{i}.1" in key]

            resnet_0_paths = renew_resnet_paths(resnets)
            paths = renew_resnet_paths(resnets)

            meta_path = {"old": f"output_blocks.{i}.0", "new": f"up_blocks.{block_id}.resnets.{layer_in_block_id}"}
            assign_to_checkpoint(
                paths, new_checkpoint, unet_state_dict, additional_replacements=[meta_path], config=config
            )

            output_block_list = {k: sorted(v) for k, v in output_block_list.items()}
            if ["conv.bias", "conv.weight"] in output_block_list.values():
                index = list(output_block_list.values()).index(["conv.bias", "conv.weight"])
                new_checkpoint[f"up_blocks.{block_id}.upsamplers.0.conv.weight"] = unet_state_dict[
                    f"output_blocks.{i}.{index}.conv.weight"
                ]
                new_checkpoint[f"up_blocks.{block_id}.upsamplers.0.conv.bias"] = unet_state_dict[
                    f"output_blocks.{i}.{index}.conv.bias"
                ]

                # Clear attentions as they have been attributed above.
                if len(attentions) == 2:
                    attentions = []

            if len(attentions):
                paths = renew_attention_paths(attentions)
                meta_path = {
                    "old": f"output_blocks.{i}.1",
                    "new": f"up_blocks.{block_id}.attentions.{layer_in_block_id}",
                }
                assign_to_checkpoint(
                    paths, new_checkpoint, unet_state_dict, additional_replacements=[meta_path], config=config
                )
        else:
            resnet_0_paths = renew_resnet_paths(output_block_layers, n_shave_prefix_segments=1)
            for path in resnet_0_paths:
                old_path = ".".join(["output_blocks", str(i), path["old"]])
                new_path = ".".join(["up_blocks", str(block_id), "resnets", str(layer_in_block_id), path["new"]])

                new_checkpoint[new_path] = unet_state_dict[old_path]

    return new_checkpoint


# Copied from diffusers.pipelines.stable_diffusion.convert_from_ckpt.convert_ldm_vae_checkpoint
def convert_ldm_vae_checkpoint(checkpoint, config):
    # extract state dict for VAE
    vae_state_dict = {}
    vae_key = "first_stage_model."
    keys = list(checkpoint.keys())
    for key in keys:
        if key.startswith(vae_key):
            vae_state_dict[key.replace(vae_key, "")] = checkpoint.get(key)

    new_checkpoint = {}

    new_checkpoint["encoder.conv_in.weight"] = vae_state_dict["encoder.conv_in.weight"]
    new_checkpoint["encoder.conv_in.bias"] = vae_state_dict["encoder.conv_in.bias"]
    new_checkpoint["encoder.conv_out.weight"] = vae_state_dict["encoder.conv_out.weight"]
    new_checkpoint["encoder.conv_out.bias"] = vae_state_dict["encoder.conv_out.bias"]
    new_checkpoint["encoder.conv_norm_out.weight"] = vae_state_dict["encoder.norm_out.weight"]
    new_checkpoint["encoder.conv_norm_out.bias"] = vae_state_dict["encoder.norm_out.bias"]

    new_checkpoint["decoder.conv_in.weight"] = vae_state_dict["decoder.conv_in.weight"]
    new_checkpoint["decoder.conv_in.bias"] = vae_state_dict["decoder.conv_in.bias"]
    new_checkpoint["decoder.conv_out.weight"] = vae_state_dict["decoder.conv_out.weight"]
    new_checkpoint["decoder.conv_out.bias"] = vae_state_dict["decoder.conv_out.bias"]
    new_checkpoint["decoder.conv_norm_out.weight"] = vae_state_dict["decoder.norm_out.weight"]
    new_checkpoint["decoder.conv_norm_out.bias"] = vae_state_dict["decoder.norm_out.bias"]

    new_checkpoint["quant_conv.weight"] = vae_state_dict["quant_conv.weight"]
    new_checkpoint["quant_conv.bias"] = vae_state_dict["quant_conv.bias"]
    new_checkpoint["post_quant_conv.weight"] = vae_state_dict["post_quant_conv.weight"]
    new_checkpoint["post_quant_conv.bias"] = vae_state_dict["post_quant_conv.bias"]

    # Retrieves the keys for the encoder down blocks only
    num_down_blocks = len({".".join(layer.split(".")[:3]) for layer in vae_state_dict if "encoder.down" in layer})
    down_blocks = {
        layer_id: [key for key in vae_state_dict if f"down.{layer_id}" in key] for layer_id in range(num_down_blocks)
    }

    # Retrieves the keys for the decoder up blocks only
    num_up_blocks = len({".".join(layer.split(".")[:3]) for layer in vae_state_dict if "decoder.up" in layer})
    up_blocks = {
        layer_id: [key for key in vae_state_dict if f"up.{layer_id}" in key] for layer_id in range(num_up_blocks)
    }

    for i in range(num_down_blocks):
        resnets = [key for key in down_blocks[i] if f"down.{i}" in key and f"down.{i}.downsample" not in key]

        if f"encoder.down.{i}.downsample.conv.weight" in vae_state_dict:
            new_checkpoint[f"encoder.down_blocks.{i}.downsamplers.0.conv.weight"] = vae_state_dict.pop(
                f"encoder.down.{i}.downsample.conv.weight"
            )
            new_checkpoint[f"encoder.down_blocks.{i}.downsamplers.0.conv.bias"] = vae_state_dict.pop(
                f"encoder.down.{i}.downsample.conv.bias"
            )

        paths = renew_vae_resnet_paths(resnets)
        meta_path = {"old": f"down.{i}.block", "new": f"down_blocks.{i}.resnets"}
        assign_to_checkpoint(paths, new_checkpoint, vae_state_dict, additional_replacements=[meta_path], config=config)

    mid_resnets = [key for key in vae_state_dict if "encoder.mid.block" in key]
    num_mid_res_blocks = 2
    for i in range(1, num_mid_res_blocks + 1):
        resnets = [key for key in mid_resnets if f"encoder.mid.block_{i}" in key]

        paths = renew_vae_resnet_paths(resnets)
        meta_path = {"old": f"mid.block_{i}", "new": f"mid_block.resnets.{i - 1}"}
        assign_to_checkpoint(paths, new_checkpoint, vae_state_dict, additional_replacements=[meta_path], config=config)

    mid_attentions = [key for key in vae_state_dict if "encoder.mid.attn" in key]
    paths = renew_vae_attention_paths(mid_attentions)
    meta_path = {"old": "mid.attn_1", "new": "mid_block.attentions.0"}
    assign_to_checkpoint(paths, new_checkpoint, vae_state_dict, additional_replacements=[meta_path], config=config)
    conv_attn_to_linear(new_checkpoint)

    for i in range(num_up_blocks):
        block_id = num_up_blocks - 1 - i
        resnets = [
            key for key in up_blocks[block_id] if f"up.{block_id}" in key and f"up.{block_id}.upsample" not in key
        ]

        if f"decoder.up.{block_id}.upsample.conv.weight" in vae_state_dict:
            new_checkpoint[f"decoder.up_blocks.{i}.upsamplers.0.conv.weight"] = vae_state_dict[
                f"decoder.up.{block_id}.upsample.conv.weight"
            ]
            new_checkpoint[f"decoder.up_blocks.{i}.upsamplers.0.conv.bias"] = vae_state_dict[
                f"decoder.up.{block_id}.upsample.conv.bias"
            ]

        paths = renew_vae_resnet_paths(resnets)
        meta_path = {"old": f"up.{block_id}.block", "new": f"up_blocks.{i}.resnets"}
        assign_to_checkpoint(paths, new_checkpoint, vae_state_dict, additional_replacements=[meta_path], config=config)

    mid_resnets = [key for key in vae_state_dict if "decoder.mid.block" in key]
    num_mid_res_blocks = 2
    for i in range(1, num_mid_res_blocks + 1):
        resnets = [key for key in mid_resnets if f"decoder.mid.block_{i}" in key]

        paths = renew_vae_resnet_paths(resnets)
        meta_path = {"old": f"mid.block_{i}", "new": f"mid_block.resnets.{i - 1}"}
        assign_to_checkpoint(paths, new_checkpoint, vae_state_dict, additional_replacements=[meta_path], config=config)

    mid_attentions = [key for key in vae_state_dict if "decoder.mid.attn" in key]
    paths = renew_vae_attention_paths(mid_attentions)
    meta_path = {"old": "mid.attn_1", "new": "mid_block.attentions.0"}
    assign_to_checkpoint(paths, new_checkpoint, vae_state_dict, additional_replacements=[meta_path], config=config)
    conv_attn_to_linear(new_checkpoint)
    return new_checkpoint


CLAP_KEYS_TO_MODIFY_MAPPING = {
    "text_branch": "text_model",
    "attn": "attention.self",
    "self.proj": "output.dense",
    "attention.self_mask": "attn_mask",
    "mlp.fc1": "intermediate.dense",
    "mlp.fc2": "output.dense",
    "norm1": "layernorm_before",
    "norm2": "layernorm_after",
    "bn0": "batch_norm",
}

CLAP_KEYS_TO_IGNORE = ["text_transform"]

CLAP_EXPECTED_MISSING_KEYS = ["text_model.embeddings.token_type_ids"]


def convert_open_clap_checkpoint(checkpoint):
    """
    Takes a state dict and returns a converted CLAP checkpoint.
    """
    # extract state dict for CLAP text embedding model, discarding the audio component
    model_state_dict = {}
    model_key = "cond_stage_model.model.text_"
    keys = list(checkpoint.keys())
    for key in keys:
        if key.startswith(model_key):
            model_state_dict[key.replace(model_key, "text_")] = checkpoint.get(key)

    new_checkpoint = {}

    sequential_layers_pattern = r".*sequential.(\d+).*"
    text_projection_pattern = r".*_projection.(\d+).*"

    for key, value in model_state_dict.items():
        # check if key should be ignored in mapping
        if key.split(".")[0] in CLAP_KEYS_TO_IGNORE:
            continue

        # check if any key needs to be modified
        for key_to_modify, new_key in CLAP_KEYS_TO_MODIFY_MAPPING.items():
            if key_to_modify in key:
                key = key.replace(key_to_modify, new_key)

        if re.match(sequential_layers_pattern, key):
            # replace sequential layers with list
            sequential_layer = re.match(sequential_layers_pattern, key).group(1)

            key = key.replace(f"sequential.{sequential_layer}.", f"layers.{int(sequential_layer)//3}.linear.")
        elif re.match(text_projection_pattern, key):
            projecton_layer = int(re.match(text_projection_pattern, key).group(1))

            # Because in CLAP they use `nn.Sequential`...
            transformers_projection_layer = 1 if projecton_layer == 0 else 2

            key = key.replace(f"_projection.{projecton_layer}.", f"_projection.linear{transformers_projection_layer}.")

        if "audio" and "qkv" in key:
            # split qkv into query key and value
            mixed_qkv = value
            qkv_dim = mixed_qkv.size(0) // 3

            query_layer = mixed_qkv[:qkv_dim]
            key_layer = mixed_qkv[qkv_dim : qkv_dim * 2]
            value_layer = mixed_qkv[qkv_dim * 2 :]

            new_checkpoint[key.replace("qkv", "query")] = query_layer
            new_checkpoint[key.replace("qkv", "key")] = key_layer
            new_checkpoint[key.replace("qkv", "value")] = value_layer
        else:
            new_checkpoint[key] = value

    return new_checkpoint


def create_transformers_vocoder_config(original_config):
    """
    Creates a config for transformers SpeechT5HifiGan based on the config of the vocoder model.
    """
    vocoder_params = original_config.model.params.vocoder_config.params

    config = {
        "model_in_dim": vocoder_params.num_mels,
        "sampling_rate": vocoder_params.sampling_rate,
        "upsample_initial_channel": vocoder_params.upsample_initial_channel,
        "upsample_rates": list(vocoder_params.upsample_rates),
        "upsample_kernel_sizes": list(vocoder_params.upsample_kernel_sizes),
        "resblock_kernel_sizes": list(vocoder_params.resblock_kernel_sizes),
        "resblock_dilation_sizes": [
            list(resblock_dilation) for resblock_dilation in vocoder_params.resblock_dilation_sizes
        ],
        "normalize_before": False,
    }

    return config


def convert_hifigan_checkpoint(checkpoint, config):
    """
    Takes a state dict and config, and returns a converted HiFiGAN vocoder checkpoint.
    """
    # extract state dict for vocoder
    vocoder_state_dict = {}
    vocoder_key = "first_stage_model.vocoder."
    keys = list(checkpoint.keys())
    for key in keys:
        if key.startswith(vocoder_key):
            vocoder_state_dict[key.replace(vocoder_key, "")] = checkpoint.get(key)

    # fix upsampler keys, everything else is correct already
    for i in range(len(config.upsample_rates)):
        vocoder_state_dict[f"upsampler.{i}.weight"] = vocoder_state_dict.pop(f"ups.{i}.weight")
        vocoder_state_dict[f"upsampler.{i}.bias"] = vocoder_state_dict.pop(f"ups.{i}.bias")

    if not config.normalize_before:
        # if we don't set normalize_before then these variables are unused, so we set them to their initialised values
        vocoder_state_dict["mean"] = torch.zeros(config.model_in_dim)
        vocoder_state_dict["scale"] = torch.ones(config.model_in_dim)

    return vocoder_state_dict


# Adapted from https://huggingface.co/spaces/haoheliu/audioldm-text-to-audio-generation/blob/84a0384742a22bd80c44e903e241f0623e874f1d/audioldm/utils.py#L72-L73
DEFAULT_CONFIG = {
    "model": {
        "params": {
            "linear_start": 0.0015,
            "linear_end": 0.0195,
            "timesteps": 1000,
            "channels": 8,
            "scale_by_std": True,
            "unet_config": {
                "target": "audioldm.latent_diffusion.openaimodel.UNetModel",
                "params": {
                    "extra_film_condition_dim": 512,
                    "extra_film_use_concat": True,
                    "in_channels": 8,
                    "out_channels": 8,
                    "model_channels": 128,
                    "attention_resolutions": [8, 4, 2],
                    "num_res_blocks": 2,
                    "channel_mult": [1, 2, 3, 5],
                    "num_head_channels": 32,
                },
            },
            "first_stage_config": {
                "target": "audioldm.variational_autoencoder.autoencoder.AutoencoderKL",
                "params": {
                    "embed_dim": 8,
                    "ddconfig": {
                        "z_channels": 8,
                        "resolution": 256,
                        "in_channels": 1,
                        "out_ch": 1,
                        "ch": 128,
                        "ch_mult": [1, 2, 4],
                        "num_res_blocks": 2,
                    },
                },
            },
            "vocoder_config": {
                "target": "audioldm.first_stage_model.vocoder",
                "params": {
                    "upsample_rates": [5, 4, 2, 2, 2],
                    "upsample_kernel_sizes": [16, 16, 8, 4, 4],
                    "upsample_initial_channel": 1024,
                    "resblock_kernel_sizes": [3, 7, 11],
                    "resblock_dilation_sizes": [[1, 3, 5], [1, 3, 5], [1, 3, 5]],
                    "num_mels": 64,
                    "sampling_rate": 16000,
                },
            },
        },
    },
}


def load_pipeline_from_original_audioldm_ckpt(
    checkpoint_path: str,
    original_config_file: str = None,
    image_size: int = 512,
    prediction_type: str = None,
    extract_ema: bool = False,
    scheduler_type: str = "ddim",
    num_in_channels: int = None,
    device: str = None,
    from_safetensors: bool = False,
) -> AudioLDMPipeline:
    """
    Load an AudioLDM pipeline object from a `.ckpt`/`.safetensors` file and (ideally) a `.yaml` config file.

    Although many of the arguments can be automatically inferred, some of these rely on brittle checks against the
    global step count, which will likely fail for models that have undergone further fine-tuning. Therefore, it is
    recommended that you override the default values and/or supply an `original_config_file` wherever possible.

    :param checkpoint_path: Path to `.ckpt` file. :param original_config_file: Path to `.yaml` config file
    corresponding to the original architecture.
            If `None`, will be automatically instantiated based on default values.
    :param image_size: The image size that the model was trained on. Use 512 for original AudioLDM checkpoints. :param
    prediction_type: The prediction type that the model was trained on. Use `'epsilon'` for original
            AudioLDM checkpoints.
    :param num_in_channels: The number of input channels. If `None` number of input channels will be automatically
            inferred.
    :param scheduler_type: Type of scheduler to use. Should be one of `["pndm", "lms", "heun", "euler",
            "euler-ancestral", "dpm", "ddim"]`.
    :param extract_ema: Only relevant for checkpoints that have both EMA and non-EMA weights. Whether to extract
            the EMA weights or not. Defaults to `False`. Pass `True` to extract the EMA weights. EMA weights usually
            yield higher quality images for inference. Non-EMA weights are usually better to continue fine-tuning.
    :param device: The device to use. Pass `None` to determine automatically. :param from_safetensors: If
    `checkpoint_path` is in `safetensors` format, load checkpoint with safetensors
            instead of PyTorch.
    :return: An AudioLDMPipeline object representing the passed-in `.ckpt`/`.safetensors` file.
    """

    if not is_omegaconf_available():
        raise ValueError(BACKENDS_MAPPING["omegaconf"][1])

    from omegaconf import OmegaConf

    if from_safetensors:
        if not is_safetensors_available():
            raise ValueError(BACKENDS_MAPPING["safetensors"][1])

        from safetensors import safe_open

        checkpoint = {}
        with safe_open(checkpoint_path, framework="pt", device="cpu") as f:
            for key in f.keys():
                checkpoint[key] = f.get_tensor(key)
    else:
        if device is None:
            device = "cuda" if torch.cuda.is_available() else "cpu"
            checkpoint = torch.load(checkpoint_path, map_location=device)
        else:
            checkpoint = torch.load(checkpoint_path, map_location=device)

    if "state_dict" in checkpoint:
        checkpoint = checkpoint["state_dict"]

    if original_config_file is None:
        original_config = DEFAULT_CONFIG
        original_config = OmegaConf.create(original_config)
    else:
        original_config = OmegaConf.load(original_config_file)

    if num_in_channels is not None:
        original_config["model"]["params"]["unet_config"]["params"]["in_channels"] = num_in_channels

    if (
        "parameterization" in original_config["model"]["params"]
        and original_config["model"]["params"]["parameterization"] == "v"
    ):
        if prediction_type is None:
            prediction_type = "v_prediction"
    else:
        if prediction_type is None:
            prediction_type = "epsilon"

    if image_size is None:
        image_size = 512

    num_train_timesteps = original_config.model.params.timesteps
    beta_start = original_config.model.params.linear_start
    beta_end = original_config.model.params.linear_end

    scheduler = DDIMScheduler(
        beta_end=beta_end,
        beta_schedule="scaled_linear",
        beta_start=beta_start,
        num_train_timesteps=num_train_timesteps,
        steps_offset=1,
        clip_sample=False,
        set_alpha_to_one=False,
        prediction_type=prediction_type,
    )
    # make sure scheduler works correctly with DDIM
    scheduler.register_to_config(clip_sample=False)

    if scheduler_type == "pndm":
        config = dict(scheduler.config)
        config["skip_prk_steps"] = True
        scheduler = PNDMScheduler.from_config(config)
    elif scheduler_type == "lms":
        scheduler = LMSDiscreteScheduler.from_config(scheduler.config)
    elif scheduler_type == "heun":
        scheduler = HeunDiscreteScheduler.from_config(scheduler.config)
    elif scheduler_type == "euler":
        scheduler = EulerDiscreteScheduler.from_config(scheduler.config)
    elif scheduler_type == "euler-ancestral":
        scheduler = EulerAncestralDiscreteScheduler.from_config(scheduler.config)
    elif scheduler_type == "dpm":
        scheduler = DPMSolverMultistepScheduler.from_config(scheduler.config)
    elif scheduler_type == "ddim":
        scheduler = scheduler
    else:
        raise ValueError(f"Scheduler of type {scheduler_type} doesn't exist!")

    # Convert the UNet2DModel
    unet_config = create_unet_diffusers_config(original_config, image_size=image_size)
    unet = UNet2DConditionModel(**unet_config)

    converted_unet_checkpoint = convert_ldm_unet_checkpoint(
        checkpoint, unet_config, path=checkpoint_path, extract_ema=extract_ema
    )

    unet.load_state_dict(converted_unet_checkpoint)

    # Convert the VAE model
    vae_config = create_vae_diffusers_config(original_config, checkpoint=checkpoint, image_size=image_size)
    converted_vae_checkpoint = convert_ldm_vae_checkpoint(checkpoint, vae_config)

    vae = AutoencoderKL(**vae_config)
    vae.load_state_dict(converted_vae_checkpoint)

    # Convert the text model
    # AudioLDM uses the same configuration and tokenizer as the original CLAP model
    config = ClapTextConfig.from_pretrained("laion/clap-htsat-unfused")
    tokenizer = AutoTokenizer.from_pretrained("laion/clap-htsat-unfused")

    converted_text_model = convert_open_clap_checkpoint(checkpoint)
    text_model = ClapTextModelWithProjection(config)

    missing_keys, unexpected_keys = text_model.load_state_dict(converted_text_model, strict=False)
    # we expect not to have token_type_ids in our original state dict so let's ignore them
    missing_keys = list(set(missing_keys) - set(CLAP_EXPECTED_MISSING_KEYS))

    if len(unexpected_keys) > 0:
        raise ValueError(f"Unexpected keys when loading CLAP model: {unexpected_keys}")

    if len(missing_keys) > 0:
        raise ValueError(f"Missing keys when loading CLAP model: {missing_keys}")

    # Convert the vocoder model
    vocoder_config = create_transformers_vocoder_config(original_config)
    vocoder_config = SpeechT5HifiGanConfig(**vocoder_config)
    converted_vocoder_checkpoint = convert_hifigan_checkpoint(checkpoint, vocoder_config)

    vocoder = SpeechT5HifiGan(vocoder_config)
    vocoder.load_state_dict(converted_vocoder_checkpoint)

    # Instantiate the diffusers pipeline
    pipe = AudioLDMPipeline(
        vae=vae,
        text_encoder=text_model,
        tokenizer=tokenizer,
        unet=unet,
        scheduler=scheduler,
        vocoder=vocoder,
    )

    return pipe


if __name__ == "__main__":
    parser = argparse.ArgumentParser()

    parser.add_argument(
        "--checkpoint_path", default=None, type=str, required=True, help="Path to the checkpoint to convert."
    )
    parser.add_argument(
        "--original_config_file",
        default=None,
        type=str,
        help="The YAML config file corresponding to the original architecture.",
    )
    parser.add_argument(
        "--num_in_channels",
        default=None,
        type=int,
        help="The number of input channels. If `None` number of input channels will be automatically inferred.",
    )
    parser.add_argument(
        "--scheduler_type",
        default="ddim",
        type=str,
        help="Type of scheduler to use. Should be one of ['pndm', 'lms', 'ddim', 'euler', 'euler-ancestral', 'dpm']",
    )
    parser.add_argument(
        "--image_size",
        default=None,
        type=int,
        help=("The image size that the model was trained on."),
    )
    parser.add_argument(
        "--prediction_type",
        default=None,
        type=str,
        help=("The prediction type that the model was trained on."),
    )
    parser.add_argument(
        "--extract_ema",
        action="store_true",
        help=(
            "Only relevant for checkpoints that have both EMA and non-EMA weights. Whether to extract the EMA weights"
            " or not. Defaults to `False`. Add `--extract_ema` to extract the EMA weights. EMA weights usually yield"
            " higher quality images for inference. Non-EMA weights are usually better to continue fine-tuning."
        ),
    )
    parser.add_argument(
        "--from_safetensors",
        action="store_true",
        help="If `--checkpoint_path` is in `safetensors` format, load checkpoint with safetensors instead of PyTorch.",
    )
    parser.add_argument(
        "--to_safetensors",
        action="store_true",
        help="Whether to store pipeline in safetensors format or not.",
    )
    parser.add_argument("--dump_path", default=None, type=str, required=True, help="Path to the output model.")
    parser.add_argument("--device", type=str, help="Device to use (e.g. cpu, cuda:0, cuda:1, etc.)")
    args = parser.parse_args()

    pipe = load_pipeline_from_original_audioldm_ckpt(
        checkpoint_path=args.checkpoint_path,
        original_config_file=args.original_config_file,
        image_size=args.image_size,
        prediction_type=args.prediction_type,
        extract_ema=args.extract_ema,
        scheduler_type=args.scheduler_type,
        num_in_channels=args.num_in_channels,
        from_safetensors=args.from_safetensors,
        device=args.device,
    )
    pipe.save_pretrained(args.dump_path, safe_serialization=args.to_safetensors)