FROM MOD
Browse files- configs/.ipynb_checkpoints/config-checkpoint.json +96 -0
- configs/config.json +100 -0
- configs/diffusion.yaml +49 -0
- logs/44k/D_0.pth +3 -0
- logs/44k/D_82400.pth +3 -0
- logs/44k/G_0.pth +3 -0
- logs/44k/G_82400.pth +3 -0
- logs/44k/feature_and_index.pkl +3 -0
- logs/44k/kmeans_10000.pt +3 -0
configs/.ipynb_checkpoints/config-checkpoint.json
ADDED
@@ -0,0 +1,96 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"train": {
|
3 |
+
"log_interval": 200,
|
4 |
+
"eval_interval": 200,
|
5 |
+
"seed": 1234,
|
6 |
+
"epochs": 10000,
|
7 |
+
"learning_rate": 0.0001,
|
8 |
+
"betas": [
|
9 |
+
0.8,
|
10 |
+
0.99
|
11 |
+
],
|
12 |
+
"eps": 1e-09,
|
13 |
+
"batch_size": 6,
|
14 |
+
"fp16_run": false,
|
15 |
+
"lr_decay": 0.999875,
|
16 |
+
"segment_size": 10240,
|
17 |
+
"init_lr_ratio": 1,
|
18 |
+
"warmup_epochs": 0,
|
19 |
+
"c_mel": 45,
|
20 |
+
"c_kl": 1.0,
|
21 |
+
"use_sr": true,
|
22 |
+
"max_speclen": 512,
|
23 |
+
"port": "8001",
|
24 |
+
"keep_ckpts": 20,
|
25 |
+
"all_in_mem": true
|
26 |
+
},
|
27 |
+
"data": {
|
28 |
+
"training_files": "filelists/train.txt",
|
29 |
+
"validation_files": "filelists/val.txt",
|
30 |
+
"max_wav_value": 32768.0,
|
31 |
+
"sampling_rate": 44100,
|
32 |
+
"filter_length": 2048,
|
33 |
+
"hop_length": 512,
|
34 |
+
"win_length": 2048,
|
35 |
+
"n_mel_channels": 80,
|
36 |
+
"mel_fmin": 0.0,
|
37 |
+
"mel_fmax": 22050
|
38 |
+
},
|
39 |
+
"model": {
|
40 |
+
"inter_channels": 192,
|
41 |
+
"hidden_channels": 192,
|
42 |
+
"filter_channels": 768,
|
43 |
+
"n_heads": 2,
|
44 |
+
"n_layers": 6,
|
45 |
+
"kernel_size": 3,
|
46 |
+
"p_dropout": 0.1,
|
47 |
+
"resblock": "1",
|
48 |
+
"resblock_kernel_sizes": [
|
49 |
+
3,
|
50 |
+
7,
|
51 |
+
11
|
52 |
+
],
|
53 |
+
"resblock_dilation_sizes": [
|
54 |
+
[
|
55 |
+
1,
|
56 |
+
3,
|
57 |
+
5
|
58 |
+
],
|
59 |
+
[
|
60 |
+
1,
|
61 |
+
3,
|
62 |
+
5
|
63 |
+
],
|
64 |
+
[
|
65 |
+
1,
|
66 |
+
3,
|
67 |
+
5
|
68 |
+
]
|
69 |
+
],
|
70 |
+
"upsample_rates": [
|
71 |
+
8,
|
72 |
+
8,
|
73 |
+
2,
|
74 |
+
2,
|
75 |
+
2
|
76 |
+
],
|
77 |
+
"upsample_initial_channel": 512,
|
78 |
+
"upsample_kernel_sizes": [
|
79 |
+
16,
|
80 |
+
16,
|
81 |
+
4,
|
82 |
+
4,
|
83 |
+
4
|
84 |
+
],
|
85 |
+
"n_layers_q": 3,
|
86 |
+
"use_spectral_norm": false,
|
87 |
+
"gin_channels": 768,
|
88 |
+
"ssl_dim": 768,
|
89 |
+
"n_speakers": 1,
|
90 |
+
"speech_encoder": "vec768l12",
|
91 |
+
"speaker_embedding": false
|
92 |
+
},
|
93 |
+
"spk": {
|
94 |
+
"renge": 0
|
95 |
+
}
|
96 |
+
}
|
configs/config.json
ADDED
@@ -0,0 +1,100 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"train": {
|
3 |
+
"log_interval": 200,
|
4 |
+
"eval_interval": 800,
|
5 |
+
"seed": 1234,
|
6 |
+
"epochs": 10000,
|
7 |
+
"learning_rate": 0.0001,
|
8 |
+
"betas": [
|
9 |
+
0.8,
|
10 |
+
0.99
|
11 |
+
],
|
12 |
+
"eps": 1e-09,
|
13 |
+
"batch_size": 6,
|
14 |
+
"fp16_run": false,
|
15 |
+
"lr_decay": 0.999875,
|
16 |
+
"segment_size": 10240,
|
17 |
+
"init_lr_ratio": 1,
|
18 |
+
"warmup_epochs": 0,
|
19 |
+
"c_mel": 45,
|
20 |
+
"c_kl": 1.0,
|
21 |
+
"use_sr": true,
|
22 |
+
"max_speclen": 512,
|
23 |
+
"port": "8001",
|
24 |
+
"keep_ckpts": 3,
|
25 |
+
"all_in_mem": false,
|
26 |
+
"vol_aug": true
|
27 |
+
},
|
28 |
+
"data": {
|
29 |
+
"training_files": "filelists/train.txt",
|
30 |
+
"validation_files": "filelists/val.txt",
|
31 |
+
"max_wav_value": 32768.0,
|
32 |
+
"sampling_rate": 44100,
|
33 |
+
"filter_length": 2048,
|
34 |
+
"hop_length": 512,
|
35 |
+
"win_length": 2048,
|
36 |
+
"n_mel_channels": 80,
|
37 |
+
"mel_fmin": 0.0,
|
38 |
+
"mel_fmax": 22050,
|
39 |
+
"unit_interpolate_mode": "nearest"
|
40 |
+
},
|
41 |
+
"model": {
|
42 |
+
"inter_channels": 192,
|
43 |
+
"hidden_channels": 192,
|
44 |
+
"filter_channels": 768,
|
45 |
+
"n_heads": 2,
|
46 |
+
"n_layers": 6,
|
47 |
+
"kernel_size": 3,
|
48 |
+
"p_dropout": 0.1,
|
49 |
+
"resblock": "1",
|
50 |
+
"resblock_kernel_sizes": [
|
51 |
+
3,
|
52 |
+
7,
|
53 |
+
11
|
54 |
+
],
|
55 |
+
"resblock_dilation_sizes": [
|
56 |
+
[
|
57 |
+
1,
|
58 |
+
3,
|
59 |
+
5
|
60 |
+
],
|
61 |
+
[
|
62 |
+
1,
|
63 |
+
3,
|
64 |
+
5
|
65 |
+
],
|
66 |
+
[
|
67 |
+
1,
|
68 |
+
3,
|
69 |
+
5
|
70 |
+
]
|
71 |
+
],
|
72 |
+
"upsample_rates": [
|
73 |
+
8,
|
74 |
+
8,
|
75 |
+
2,
|
76 |
+
2,
|
77 |
+
2
|
78 |
+
],
|
79 |
+
"upsample_initial_channel": 512,
|
80 |
+
"upsample_kernel_sizes": [
|
81 |
+
16,
|
82 |
+
16,
|
83 |
+
4,
|
84 |
+
4,
|
85 |
+
4
|
86 |
+
],
|
87 |
+
"n_layers_q": 3,
|
88 |
+
"use_spectral_norm": false,
|
89 |
+
"gin_channels": 768,
|
90 |
+
"ssl_dim": 768,
|
91 |
+
"n_speakers": 1,
|
92 |
+
"vocoder_name": "nsf-hifigan",
|
93 |
+
"speech_encoder": "vec768l12",
|
94 |
+
"speaker_embedding": false,
|
95 |
+
"vol_embedding": true
|
96 |
+
},
|
97 |
+
"spk": {
|
98 |
+
"1001": 0
|
99 |
+
}
|
100 |
+
}
|
configs/diffusion.yaml
ADDED
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
data:
|
2 |
+
block_size: 512
|
3 |
+
cnhubertsoft_gate: 10
|
4 |
+
duration: 2
|
5 |
+
encoder: vec768l12
|
6 |
+
encoder_hop_size: 320
|
7 |
+
encoder_out_channels: 768
|
8 |
+
encoder_sample_rate: 16000
|
9 |
+
extensions:
|
10 |
+
- wav
|
11 |
+
sampling_rate: 44100
|
12 |
+
training_files: filelists/train.txt
|
13 |
+
unit_interpolate_mode: nearest
|
14 |
+
validation_files: filelists/val.txt
|
15 |
+
device: cuda
|
16 |
+
env:
|
17 |
+
expdir: logs/44k/diffusion
|
18 |
+
gpu_id: 0
|
19 |
+
infer:
|
20 |
+
method: dpm-solver
|
21 |
+
speedup: 10
|
22 |
+
model:
|
23 |
+
n_chans: 512
|
24 |
+
n_hidden: 256
|
25 |
+
n_layers: 20
|
26 |
+
n_spk: 1
|
27 |
+
type: Diffusion
|
28 |
+
use_pitch_aug: true
|
29 |
+
spk:
|
30 |
+
'1001': 0
|
31 |
+
train:
|
32 |
+
amp_dtype: fp32
|
33 |
+
batch_size: 48
|
34 |
+
cache_all_data: true
|
35 |
+
cache_device: cpu
|
36 |
+
cache_fp16: true
|
37 |
+
decay_step: 100000
|
38 |
+
epochs: 100000
|
39 |
+
gamma: 0.5
|
40 |
+
interval_force_save: 10000
|
41 |
+
interval_log: 10
|
42 |
+
interval_val: 2000
|
43 |
+
lr: 0.0002
|
44 |
+
num_workers: 2
|
45 |
+
save_opt: false
|
46 |
+
weight_decay: 0
|
47 |
+
vocoder:
|
48 |
+
ckpt: pretrain/nsf_hifigan/model
|
49 |
+
type: nsf-hifigan
|
logs/44k/D_0.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2506a5457b2f6133563418ab0429463d3ca9788cc2575979d3839a8699b3c158
|
3 |
+
size 187027770
|
logs/44k/D_82400.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4190317d45788a4592202e1a4b60037efb431b1d5b63c0e7a94567c2f372712a
|
3 |
+
size 561098185
|
logs/44k/G_0.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:176d6e0dde4ba7de5ebfe1d01a1afebb161ec9945ea5cbee3131f187f53b39be
|
3 |
+
size 209270847
|
logs/44k/G_82400.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0b6c7cbbfef01b7df606c9b72d18850fa4c22065fc7d436b652bf7add06ff2c8
|
3 |
+
size 627912085
|
logs/44k/feature_and_index.pkl
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5907578cb02a14cf69d91c0ba65eb52642e670495662a24c6491aab5dff2b4ca
|
3 |
+
size 981873425
|
logs/44k/kmeans_10000.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ae43ee23470208cccb34a090acd2863428e682018e0e9e0ff13ccab795931868
|
3 |
+
size 33164409
|