Hugging Face's logo

chameleon-ai
/
so-vits-svc-models

Audio-to-Audio
so-vits-svc
voice-cloning
Model card Files Community
chameleon-ai commited on
Commit
e2b3455
•
1 Parent(s): a4fddb7

Upload 4 files

Browse files
Files changed (4) hide show
  1. edi/G_800.pth +3 -0
  2. edi/config.json +98 -0
  3. ezio-auditore/G_507.pth +3 -0
  4. ezio-auditore/config.json +104 -0
edi/G_800.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1084701e017605a5295ec75b2d268ba4b0c2d63df0a780f937595d9e8fcb817f
3
+ size 542789469
edi/config.json ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 800,
5
+ "seed": 1234,
6
+ "epochs": 400,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 16,
14
+ "fp16_run": false,
15
+ "bf16_run": false,
16
+ "lr_decay": 0.999875,
17
+ "segment_size": 10240,
18
+ "init_lr_ratio": 1,
19
+ "warmup_epochs": 0,
20
+ "c_mel": 45,
21
+ "c_kl": 1.0,
22
+ "use_sr": true,
23
+ "max_speclen": 512,
24
+ "port": "8001",
25
+ "keep_ckpts": 20,
26
+ "num_workers": 4,
27
+ "log_version": 0,
28
+ "ckpt_name_by_step": false,
29
+ "accumulate_grad_batches": 1
30
+ },
31
+ "data": {
32
+ "training_files": "filelists/44k/train.txt",
33
+ "validation_files": "filelists/44k/val.txt",
34
+ "max_wav_value": 32768.0,
35
+ "sampling_rate": 44100,
36
+ "filter_length": 2048,
37
+ "hop_length": 512,
38
+ "win_length": 2048,
39
+ "n_mel_channels": 80,
40
+ "mel_fmin": 0.0,
41
+ "mel_fmax": 22050
42
+ },
43
+ "model": {
44
+ "inter_channels": 192,
45
+ "hidden_channels": 192,
46
+ "filter_channels": 768,
47
+ "n_heads": 2,
48
+ "n_layers": 6,
49
+ "kernel_size": 3,
50
+ "p_dropout": 0.1,
51
+ "resblock": "1",
52
+ "resblock_kernel_sizes": [
53
+ 3,
54
+ 7,
55
+ 11
56
+ ],
57
+ "resblock_dilation_sizes": [
58
+ [
59
+ 1,
60
+ 3,
61
+ 5
62
+ ],
63
+ [
64
+ 1,
65
+ 3,
66
+ 5
67
+ ],
68
+ [
69
+ 1,
70
+ 3,
71
+ 5
72
+ ]
73
+ ],
74
+ "upsample_rates": [
75
+ 8,
76
+ 8,
77
+ 2,
78
+ 2,
79
+ 2
80
+ ],
81
+ "upsample_initial_channel": 512,
82
+ "upsample_kernel_sizes": [
83
+ 16,
84
+ 16,
85
+ 4,
86
+ 4,
87
+ 4
88
+ ],
89
+ "n_layers_q": 3,
90
+ "use_spectral_norm": false,
91
+ "gin_channels": 256,
92
+ "ssl_dim": 256,
93
+ "n_speakers": 200
94
+ },
95
+ "spk": {
96
+ "edi": 0
97
+ }
98
+ }
ezio-auditore/G_507.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86b769b05c0f91df20bc4b2bec02c44bdeca54726552a4585339399e68e20fc0
3
+ size 548687709
ezio-auditore/config.json ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 100,
4
+ "eval_interval": 200,
5
+ "seed": 1234,
6
+ "epochs": 800,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 16,
14
+ "fp16_run": false,
15
+ "bf16_run": false,
16
+ "lr_decay": 0.999875,
17
+ "segment_size": 10240,
18
+ "init_lr_ratio": 1,
19
+ "warmup_epochs": 0,
20
+ "c_mel": 45,
21
+ "c_kl": 1.0,
22
+ "use_sr": true,
23
+ "max_speclen": 512,
24
+ "port": "8001",
25
+ "keep_ckpts": 25,
26
+ "num_workers": 4,
27
+ "log_version": 0,
28
+ "ckpt_name_by_step": false,
29
+ "accumulate_grad_batches": 1
30
+ },
31
+ "data": {
32
+ "training_files": "filelists/44k/train.txt",
33
+ "validation_files": "filelists/44k/val.txt",
34
+ "max_wav_value": 32768.0,
35
+ "sampling_rate": 44100,
36
+ "filter_length": 2048,
37
+ "hop_length": 512,
38
+ "win_length": 2048,
39
+ "n_mel_channels": 80,
40
+ "mel_fmin": 0.0,
41
+ "mel_fmax": 22050,
42
+ "contentvec_final_proj": false
43
+ },
44
+ "model": {
45
+ "inter_channels": 192,
46
+ "hidden_channels": 192,
47
+ "filter_channels": 768,
48
+ "n_heads": 2,
49
+ "n_layers": 6,
50
+ "kernel_size": 3,
51
+ "p_dropout": 0.1,
52
+ "resblock": "1",
53
+ "resblock_kernel_sizes": [
54
+ 3,
55
+ 7,
56
+ 11
57
+ ],
58
+ "resblock_dilation_sizes": [
59
+ [
60
+ 1,
61
+ 3,
62
+ 5
63
+ ],
64
+ [
65
+ 1,
66
+ 3,
67
+ 5
68
+ ],
69
+ [
70
+ 1,
71
+ 3,
72
+ 5
73
+ ]
74
+ ],
75
+ "upsample_rates": [
76
+ 8,
77
+ 8,
78
+ 2,
79
+ 2,
80
+ 2
81
+ ],
82
+ "upsample_initial_channel": 512,
83
+ "upsample_kernel_sizes": [
84
+ 16,
85
+ 16,
86
+ 4,
87
+ 4,
88
+ 4
89
+ ],
90
+ "n_layers_q": 3,
91
+ "use_spectral_norm": false,
92
+ "gin_channels": 256,
93
+ "ssl_dim": 768,
94
+ "n_speakers": 200,
95
+ "type_": "hifi-gan",
96
+ "pretrained": {
97
+ "D_0.pth": "https://huggingface.co/datasets/ms903/sovits4.0-768vec-layer12/resolve/main/sovits_768l12_pre_large_320k/clean_D_320000.pth",
98
+ "G_0.pth": "https://huggingface.co/datasets/ms903/sovits4.0-768vec-layer12/resolve/main/sovits_768l12_pre_large_320k/clean_G_320000.pth"
99
+ }
100
+ },
101
+ "spk": {
102
+ "ezio-auditore": 0
103
+ }
104
+ }