Commit
·
7039efb
1
Parent(s):
b75291e
Upload 9 files
Browse files- Chtholly_V5Co-1076epoch-80800step-Vec768-Layer12.pth +3 -0
- Chtholly_V5Co-1076epoch-80800step-Vec768-Layer12_compressed.pth +3 -0
- Chtholly_V5Sp-527epoch-21600step-Vec768-Layer12.pth +3 -0
- Chtholly_V5Sp-527epoch-21600step-Vec768-Layer12_compressed.pth +3 -0
- Chtholly_V5_config.json +96 -0
- Chtholly_V5_config.yaml +48 -0
- Chtholly_V5_kmeans_10000.pt +3 -0
- Chtholly_V5_model_52000.pt +3 -0
- /345/243/260/347/272/271/347/233/270/344/274/274/345/272/246.png +0 -0
Chtholly_V5Co-1076epoch-80800step-Vec768-Layer12.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b054ad03b0822227050fac368da6602563c1500ee38dcb2c4df5e969801f6abd
|
| 3 |
+
size 627905309
|
Chtholly_V5Co-1076epoch-80800step-Vec768-Layer12_compressed.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6dd4464efe9f017c44bdfbd77073279bafec9ad1d17839ff402deb89e8cfb3d9
|
| 3 |
+
size 209255614
|
Chtholly_V5Sp-527epoch-21600step-Vec768-Layer12.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:591324a80369d7a768d93eb448bc059f1577c1277cc0e3644ed20eaa129c3531
|
| 3 |
+
size 627905309
|
Chtholly_V5Sp-527epoch-21600step-Vec768-Layer12_compressed.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8bef984ab6e7cfa46996e26fb1523705cee692ac0478b8bfc1c0088cc4e18466
|
| 3 |
+
size 209255614
|
Chtholly_V5_config.json
ADDED
|
@@ -0,0 +1,96 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 12,
|
| 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": 10,
|
| 25 |
+
"all_in_mem": false
|
| 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 |
+
"Chtholly_V5": 0
|
| 95 |
+
}
|
| 96 |
+
}
|
Chtholly_V5_config.yaml
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
data:
|
| 2 |
+
block_size: 512
|
| 3 |
+
cnhubertsoft_gate: 10
|
| 4 |
+
duration: 1
|
| 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 |
+
validation_files: filelists/val.txt
|
| 14 |
+
device: cuda
|
| 15 |
+
env:
|
| 16 |
+
expdir: logs/44k/diffusion
|
| 17 |
+
gpu_id: 0
|
| 18 |
+
infer:
|
| 19 |
+
method: dpm-solver
|
| 20 |
+
speedup: 10
|
| 21 |
+
model:
|
| 22 |
+
n_chans: 512
|
| 23 |
+
n_hidden: 256
|
| 24 |
+
n_layers: 20
|
| 25 |
+
n_spk: 1
|
| 26 |
+
type: Diffusion
|
| 27 |
+
use_pitch_aug: true
|
| 28 |
+
spk:
|
| 29 |
+
Chtholly_V5: 0
|
| 30 |
+
train:
|
| 31 |
+
amp_dtype: fp32
|
| 32 |
+
batch_size: 48
|
| 33 |
+
cache_all_data: true
|
| 34 |
+
cache_device: cpu
|
| 35 |
+
cache_fp16: true
|
| 36 |
+
decay_step: 100000
|
| 37 |
+
epochs: 100000
|
| 38 |
+
gamma: 0.5
|
| 39 |
+
interval_force_save: 10000
|
| 40 |
+
interval_log: 10
|
| 41 |
+
interval_val: 2000
|
| 42 |
+
lr: 0.0002
|
| 43 |
+
num_workers: 2
|
| 44 |
+
save_opt: false
|
| 45 |
+
weight_decay: 0
|
| 46 |
+
vocoder:
|
| 47 |
+
ckpt: pretrain/nsf_hifigan/model
|
| 48 |
+
type: nsf-hifigan
|
Chtholly_V5_kmeans_10000.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:788ae6d94f953764ca3605bf4996f2ee61964058d4b1d858624ed50446d08168
|
| 3 |
+
size 32539129
|
Chtholly_V5_model_52000.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0cfd0b35c8049b380107417597c4029c33c1e81f9593973e1cee078ebd3db3ef
|
| 3 |
+
size 220893767
|
/345/243/260/347/272/271/347/233/270/344/274/274/345/272/246.png
ADDED
|