AzumaSeren100 commited on
Commit
b823bbf
1 Parent(s): 554fcfd

Upload 2 files

Browse files
logs/xuanshen/G_26800.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17bfbec17a3592669a8350e9b5dcb26e7c949e5ede02fb876d915975f2d0b38b
3
+ size 621696821
logs/xuanshen/config.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 100,
4
+ "eval_interval": 200,
5
+ "seed": 52,
6
+ "epochs": 10000,
7
+ "learning_rate": 0.00015,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 12,
14
+ "fp16_run": false,
15
+ "lr_decay": 0.999875,
16
+ "segment_size": 16384,
17
+ "init_lr_ratio": 1,
18
+ "warmup_epochs": 0,
19
+ "c_mel": 45,
20
+ "c_kl": 1.0
21
+ },
22
+ "data": {
23
+ "use_mel_posterior_encoder": false,
24
+ "training_files": "filelists/train.list",
25
+ "validation_files": "filelists/val.list",
26
+ "max_wav_value": 32768.0,
27
+ "sampling_rate": 44100,
28
+ "filter_length": 2048,
29
+ "hop_length": 512,
30
+ "win_length": 2048,
31
+ "n_mel_channels": 128,
32
+ "mel_fmin": 0.0,
33
+ "mel_fmax": null,
34
+ "add_blank": true,
35
+ "n_speakers": 256,
36
+ "cleaned_text": true,
37
+ "spk2id": {
38
+ "xuanshen": 0,
39
+ "dxl": 248
40
+ }
41
+ },
42
+ "model": {
43
+ "use_spk_conditioned_encoder": true,
44
+ "use_noise_scaled_mas": true,
45
+ "use_mel_posterior_encoder": false,
46
+ "use_duration_discriminator": true,
47
+ "inter_channels": 192,
48
+ "hidden_channels": 192,
49
+ "filter_channels": 768,
50
+ "n_heads": 2,
51
+ "n_layers": 6,
52
+ "kernel_size": 3,
53
+ "p_dropout": 0.1,
54
+ "resblock": "1",
55
+ "resblock_kernel_sizes": [
56
+ 3,
57
+ 7,
58
+ 11
59
+ ],
60
+ "resblock_dilation_sizes": [
61
+ [
62
+ 1,
63
+ 3,
64
+ 5
65
+ ],
66
+ [
67
+ 1,
68
+ 3,
69
+ 5
70
+ ],
71
+ [
72
+ 1,
73
+ 3,
74
+ 5
75
+ ]
76
+ ],
77
+ "upsample_rates": [
78
+ 8,
79
+ 8,
80
+ 2,
81
+ 2,
82
+ 2
83
+ ],
84
+ "upsample_initial_channel": 512,
85
+ "upsample_kernel_sizes": [
86
+ 16,
87
+ 16,
88
+ 8,
89
+ 2,
90
+ 2
91
+ ],
92
+ "n_layers_q": 3,
93
+ "use_spectral_norm": false,
94
+ "gin_channels": 256
95
+ }
96
+ }