Kit-Lemonfoot
commited on
Commit
•
926eb83
1
Parent(s):
af54be2
Added Pippa and Finana, fixed a common VITS deprecation issue to aid in logs declutter
Browse files- .gitattributes +10 -0
- GPT_SoVITS/GPT_weights/FinanaRyugu_GPT.ckpt +3 -0
- GPT_SoVITS/GPT_weights/PipkinPippa_GPT.ckpt +3 -0
- GPT_SoVITS/SoVITS_weights/FinanaRyugu_SoVITS.pth +3 -0
- GPT_SoVITS/SoVITS_weights/PipkinPippa_SoVITS.pth +3 -0
- GPT_SoVITS/TTS_infer_pack/TTS.py +4 -4
- GPT_SoVITS/TTS_infer_pack/TextPreprocessor.py +5 -5
- GPT_SoVITS/inference_webui.py +0 -1
- GPT_SoVITS/module/attentions.py +2 -1
- GPT_SoVITS/module/models.py +2 -1
- GPT_SoVITS/module/models_onnx.py +2 -1
- GPT_SoVITS/module/modules.py +5 -4
- GPT_SoVITS/module/mrte_model.py +2 -1
- images/finana.png +0 -0
- images/pippa.png +0 -0
- referenceaudio/Finana/A1.wav +3 -0
- referenceaudio/Finana/A2.wav +3 -0
- referenceaudio/Finana/A3.wav +3 -0
- referenceaudio/Finana/A4.wav +3 -0
- referenceaudio/Finana/A5.wav +3 -0
- referenceaudio/Pippa/A1.wav +3 -0
- referenceaudio/Pippa/A2.wav +3 -0
- referenceaudio/Pippa/A3.wav +3 -0
- referenceaudio/Pippa/A4.wav +3 -0
- referenceaudio/Pippa/A5.wav +3 -0
- voicelist.json +183 -139
.gitattributes
CHANGED
@@ -49,3 +49,13 @@ referenceaudio/Dokibird/A4.wav filter=lfs diff=lfs merge=lfs -text
|
|
49 |
referenceaudio/Dokibird/A5.wav filter=lfs diff=lfs merge=lfs -text
|
50 |
referenceaudio/Dokibird/A6.wav filter=lfs diff=lfs merge=lfs -text
|
51 |
referenceaudio/Dokibird/A7.wav filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
49 |
referenceaudio/Dokibird/A5.wav filter=lfs diff=lfs merge=lfs -text
|
50 |
referenceaudio/Dokibird/A6.wav filter=lfs diff=lfs merge=lfs -text
|
51 |
referenceaudio/Dokibird/A7.wav filter=lfs diff=lfs merge=lfs -text
|
52 |
+
referenceaudio/Finana/A1.wav filter=lfs diff=lfs merge=lfs -text
|
53 |
+
referenceaudio/Finana/A2.wav filter=lfs diff=lfs merge=lfs -text
|
54 |
+
referenceaudio/Finana/A3.wav filter=lfs diff=lfs merge=lfs -text
|
55 |
+
referenceaudio/Finana/A4.wav filter=lfs diff=lfs merge=lfs -text
|
56 |
+
referenceaudio/Finana/A5.wav filter=lfs diff=lfs merge=lfs -text
|
57 |
+
referenceaudio/Pippa/A1.wav filter=lfs diff=lfs merge=lfs -text
|
58 |
+
referenceaudio/Pippa/A2.wav filter=lfs diff=lfs merge=lfs -text
|
59 |
+
referenceaudio/Pippa/A3.wav filter=lfs diff=lfs merge=lfs -text
|
60 |
+
referenceaudio/Pippa/A4.wav filter=lfs diff=lfs merge=lfs -text
|
61 |
+
referenceaudio/Pippa/A5.wav filter=lfs diff=lfs merge=lfs -text
|
GPT_SoVITS/GPT_weights/FinanaRyugu_GPT.ckpt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4004ff21d3d0a509ff5a245800adeb6c2d488d7674f7b7a05c40267b1f3ed9df
|
3 |
+
size 155087286
|
GPT_SoVITS/GPT_weights/PipkinPippa_GPT.ckpt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e8cb6b73f207dd43ae687aefe544d1917c1e75c9b7efc86e9ae24c5f36666083
|
3 |
+
size 155087286
|
GPT_SoVITS/SoVITS_weights/FinanaRyugu_SoVITS.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c049d72d47671ece9c45f8e1c359db359462b6593bc19e12ed3cc210065d27df
|
3 |
+
size 84884104
|
GPT_SoVITS/SoVITS_weights/PipkinPippa_SoVITS.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:575877890fabc2d91386c3e5790bbce6c5ca15be6691e7c150c619285fb659c4
|
3 |
+
size 84885454
|
GPT_SoVITS/TTS_infer_pack/TTS.py
CHANGED
@@ -52,7 +52,7 @@ custom:
|
|
52 |
def set_seed(seed:int):
|
53 |
seed = int(seed)
|
54 |
seed = seed if seed != -1 else random.randrange(1 << 32)
|
55 |
-
print(f"
|
56 |
os.environ['PYTHONHASHSEED'] = str(seed)
|
57 |
random.seed(seed)
|
58 |
np.random.seed(seed)
|
@@ -229,7 +229,7 @@ class TTS:
|
|
229 |
|
230 |
|
231 |
def init_cnhuhbert_weights(self, base_path: str):
|
232 |
-
print(f"Loading CNHuBERT weights from {base_path}")
|
233 |
self.cnhuhbert_model = CNHubert(base_path)
|
234 |
self.cnhuhbert_model=self.cnhuhbert_model.eval()
|
235 |
self.cnhuhbert_model = self.cnhuhbert_model.to(self.configs.device)
|
@@ -280,7 +280,7 @@ class TTS:
|
|
280 |
|
281 |
|
282 |
def init_t2s_weights(self, weights_path: str):
|
283 |
-
print(f"Loading Text2Semantic weights from {weights_path}")
|
284 |
self.configs.t2s_weights_path = weights_path
|
285 |
self.configs.save_configs()
|
286 |
self.configs.hz = 50
|
@@ -929,4 +929,4 @@ def speed_change(input_audio:np.ndarray, speed:float, sr:int):
|
|
929 |
# 将管道输出解码为 NumPy 数组
|
930 |
processed_audio = np.frombuffer(out, np.int16)
|
931 |
|
932 |
-
return processed_audio
|
|
|
52 |
def set_seed(seed:int):
|
53 |
seed = int(seed)
|
54 |
seed = seed if seed != -1 else random.randrange(1 << 32)
|
55 |
+
print(f"Seed: {seed}")
|
56 |
os.environ['PYTHONHASHSEED'] = str(seed)
|
57 |
random.seed(seed)
|
58 |
np.random.seed(seed)
|
|
|
229 |
|
230 |
|
231 |
def init_cnhuhbert_weights(self, base_path: str):
|
232 |
+
#print(f"Loading CNHuBERT weights from {base_path}")
|
233 |
self.cnhuhbert_model = CNHubert(base_path)
|
234 |
self.cnhuhbert_model=self.cnhuhbert_model.eval()
|
235 |
self.cnhuhbert_model = self.cnhuhbert_model.to(self.configs.device)
|
|
|
280 |
|
281 |
|
282 |
def init_t2s_weights(self, weights_path: str):
|
283 |
+
#print(f"Loading Text2Semantic weights from {weights_path}")
|
284 |
self.configs.t2s_weights_path = weights_path
|
285 |
self.configs.save_configs()
|
286 |
self.configs.hz = 50
|
|
|
929 |
# 将管道输出解码为 NumPy 数组
|
930 |
processed_audio = np.frombuffer(out, np.int16)
|
931 |
|
932 |
+
return processed_audio
|
GPT_SoVITS/TTS_infer_pack/TextPreprocessor.py
CHANGED
@@ -53,10 +53,10 @@ class TextPreprocessor:
|
|
53 |
self.device = device
|
54 |
|
55 |
def preprocess(self, text:str, lang:str, text_split_method:str)->List[Dict]:
|
56 |
-
print(i18n("############ 切分文本 ############"))
|
57 |
texts = self.pre_seg_text(text, lang, text_split_method)
|
58 |
result = []
|
59 |
-
print(i18n("############ 提取文本Bert特征 ############"))
|
60 |
for text in tqdm(texts):
|
61 |
phones, bert_features, norm_text = self.segment_and_extract_feature_for_text(text, lang)
|
62 |
if phones is None:
|
@@ -73,8 +73,8 @@ class TextPreprocessor:
|
|
73 |
text = text.strip("\n")
|
74 |
if (text[0] not in splits and len(get_first(text)) < 4):
|
75 |
text = "。" + text if lang != "en" else "." + text
|
76 |
-
print(i18n("实际输入的目标文本:"))
|
77 |
-
print(text)
|
78 |
|
79 |
seg_method = get_seg_method(text_split_method)
|
80 |
text = seg_method(text)
|
@@ -99,7 +99,7 @@ class TextPreprocessor:
|
|
99 |
else:
|
100 |
texts.append(text)
|
101 |
|
102 |
-
print(
|
103 |
print(texts)
|
104 |
return texts
|
105 |
|
|
|
53 |
self.device = device
|
54 |
|
55 |
def preprocess(self, text:str, lang:str, text_split_method:str)->List[Dict]:
|
56 |
+
#print(i18n("############ 切分文本 ############"))
|
57 |
texts = self.pre_seg_text(text, lang, text_split_method)
|
58 |
result = []
|
59 |
+
#print(i18n("############ 提取文本Bert特征 ############"))
|
60 |
for text in tqdm(texts):
|
61 |
phones, bert_features, norm_text = self.segment_and_extract_feature_for_text(text, lang)
|
62 |
if phones is None:
|
|
|
73 |
text = text.strip("\n")
|
74 |
if (text[0] not in splits and len(get_first(text)) < 4):
|
75 |
text = "。" + text if lang != "en" else "." + text
|
76 |
+
#print(i18n("实际输入的目标文本:"))
|
77 |
+
#print(text)
|
78 |
|
79 |
seg_method = get_seg_method(text_split_method)
|
80 |
text = seg_method(text)
|
|
|
99 |
else:
|
100 |
texts.append(text)
|
101 |
|
102 |
+
print("Starting inference on the following text:")
|
103 |
print(texts)
|
104 |
return texts
|
105 |
|
GPT_SoVITS/inference_webui.py
CHANGED
@@ -95,7 +95,6 @@ def inference(name, gptmp, svmp, sty, text, text_lang,
|
|
95 |
if clm!=name:
|
96 |
print(f"Switching to model {name}")
|
97 |
clm=name
|
98 |
-
print(os.getcwd())
|
99 |
tts_pipeline.init_t2s_weights(gptmp)
|
100 |
tts_pipeline.init_vits_weights(svmp)
|
101 |
|
|
|
95 |
if clm!=name:
|
96 |
print(f"Switching to model {name}")
|
97 |
clm=name
|
|
|
98 |
tts_pipeline.init_t2s_weights(gptmp)
|
99 |
tts_pipeline.init_vits_weights(svmp)
|
100 |
|
GPT_SoVITS/module/attentions.py
CHANGED
@@ -433,7 +433,8 @@ class FFN(nn.Module):
|
|
433 |
|
434 |
|
435 |
import torch.nn as nn
|
436 |
-
from torch.nn.utils import remove_weight_norm
|
|
|
437 |
|
438 |
|
439 |
class Depthwise_Separable_Conv1D(nn.Module):
|
|
|
433 |
|
434 |
|
435 |
import torch.nn as nn
|
436 |
+
from torch.nn.utils import remove_weight_norm
|
437 |
+
from torch.nn.utils.parametrizations import weight_norm
|
438 |
|
439 |
|
440 |
class Depthwise_Separable_Conv1D(nn.Module):
|
GPT_SoVITS/module/models.py
CHANGED
@@ -10,7 +10,8 @@ from module import modules
|
|
10 |
from module import attentions
|
11 |
|
12 |
from torch.nn import Conv1d, ConvTranspose1d, AvgPool1d, Conv2d
|
13 |
-
from torch.nn.utils import
|
|
|
14 |
from module.commons import init_weights, get_padding
|
15 |
from module.mrte_model import MRTE
|
16 |
from module.quantize import ResidualVectorQuantizer
|
|
|
10 |
from module import attentions
|
11 |
|
12 |
from torch.nn import Conv1d, ConvTranspose1d, AvgPool1d, Conv2d
|
13 |
+
from torch.nn.utils import remove_weight_norm, spectral_norm
|
14 |
+
from torch.nn.utils.parametrizations import weight_norm
|
15 |
from module.commons import init_weights, get_padding
|
16 |
from module.mrte_model import MRTE
|
17 |
from module.quantize import ResidualVectorQuantizer
|
GPT_SoVITS/module/models_onnx.py
CHANGED
@@ -9,7 +9,8 @@ from module import modules
|
|
9 |
from module import attentions_onnx as attentions
|
10 |
|
11 |
from torch.nn import Conv1d, ConvTranspose1d, AvgPool1d, Conv2d
|
12 |
-
from torch.nn.utils import
|
|
|
13 |
from module.commons import init_weights, get_padding
|
14 |
from module.mrte_model import MRTE
|
15 |
from module.quantize import ResidualVectorQuantizer
|
|
|
9 |
from module import attentions_onnx as attentions
|
10 |
|
11 |
from torch.nn import Conv1d, ConvTranspose1d, AvgPool1d, Conv2d
|
12 |
+
from torch.nn.utils import remove_weight_norm, spectral_norm
|
13 |
+
from torch.nn.utils.parametrizations import weight_norm
|
14 |
from module.commons import init_weights, get_padding
|
15 |
from module.mrte_model import MRTE
|
16 |
from module.quantize import ResidualVectorQuantizer
|
GPT_SoVITS/module/modules.py
CHANGED
@@ -5,7 +5,8 @@ from torch import nn
|
|
5 |
from torch.nn import functional as F
|
6 |
|
7 |
from torch.nn import Conv1d
|
8 |
-
from torch.nn.utils import
|
|
|
9 |
|
10 |
from module import commons
|
11 |
from module.commons import init_weights, get_padding
|
@@ -159,7 +160,7 @@ class WN(torch.nn.Module):
|
|
159 |
cond_layer = torch.nn.Conv1d(
|
160 |
gin_channels, 2 * hidden_channels * n_layers, 1
|
161 |
)
|
162 |
-
self.cond_layer = torch.nn.utils.weight_norm(cond_layer, name="weight")
|
163 |
|
164 |
for i in range(n_layers):
|
165 |
dilation = dilation_rate**i
|
@@ -171,7 +172,7 @@ class WN(torch.nn.Module):
|
|
171 |
dilation=dilation,
|
172 |
padding=padding,
|
173 |
)
|
174 |
-
in_layer = torch.nn.utils.weight_norm(in_layer, name="weight")
|
175 |
self.in_layers.append(in_layer)
|
176 |
|
177 |
# last one is not necessary
|
@@ -181,7 +182,7 @@ class WN(torch.nn.Module):
|
|
181 |
res_skip_channels = hidden_channels
|
182 |
|
183 |
res_skip_layer = torch.nn.Conv1d(hidden_channels, res_skip_channels, 1)
|
184 |
-
res_skip_layer = torch.nn.utils.weight_norm(res_skip_layer, name="weight")
|
185 |
self.res_skip_layers.append(res_skip_layer)
|
186 |
|
187 |
def forward(self, x, x_mask, g=None, **kwargs):
|
|
|
5 |
from torch.nn import functional as F
|
6 |
|
7 |
from torch.nn import Conv1d
|
8 |
+
from torch.nn.utils import remove_weight_norm
|
9 |
+
from torch.nn.utils.parametrizations import weight_norm
|
10 |
|
11 |
from module import commons
|
12 |
from module.commons import init_weights, get_padding
|
|
|
160 |
cond_layer = torch.nn.Conv1d(
|
161 |
gin_channels, 2 * hidden_channels * n_layers, 1
|
162 |
)
|
163 |
+
self.cond_layer = torch.nn.utils.parametrizations.weight_norm(cond_layer, name="weight")
|
164 |
|
165 |
for i in range(n_layers):
|
166 |
dilation = dilation_rate**i
|
|
|
172 |
dilation=dilation,
|
173 |
padding=padding,
|
174 |
)
|
175 |
+
in_layer = torch.nn.utils.parametrizations.weight_norm(in_layer, name="weight")
|
176 |
self.in_layers.append(in_layer)
|
177 |
|
178 |
# last one is not necessary
|
|
|
182 |
res_skip_channels = hidden_channels
|
183 |
|
184 |
res_skip_layer = torch.nn.Conv1d(hidden_channels, res_skip_channels, 1)
|
185 |
+
res_skip_layer = torch.nn.utils.parametrizations.weight_norm(res_skip_layer, name="weight")
|
186 |
self.res_skip_layers.append(res_skip_layer)
|
187 |
|
188 |
def forward(self, x, x_mask, g=None, **kwargs):
|
GPT_SoVITS/module/mrte_model.py
CHANGED
@@ -2,7 +2,8 @@
|
|
2 |
|
3 |
import torch
|
4 |
from torch import nn
|
5 |
-
from torch.nn.utils import remove_weight_norm
|
|
|
6 |
from module.attentions import MultiHeadAttention
|
7 |
|
8 |
|
|
|
2 |
|
3 |
import torch
|
4 |
from torch import nn
|
5 |
+
from torch.nn.utils import remove_weight_norm
|
6 |
+
from torch.nn.utils.parametrizations import weight_norm
|
7 |
from module.attentions import MultiHeadAttention
|
8 |
|
9 |
|
images/finana.png
ADDED
images/pippa.png
ADDED
referenceaudio/Finana/A1.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8e49131f6f8c448ee390f74f03f63c69efdf667cdc6196c393c4738ed3e05259
|
3 |
+
size 1254440
|
referenceaudio/Finana/A2.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:14d6adb2058a5faa779f9d1fa865901f72ed7d1ba349117292445a476d692bce
|
3 |
+
size 1459062
|
referenceaudio/Finana/A3.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:da16f87d463f773272fe57f5507d822b2290a39bcdff0d63ab663f64ce4b768a
|
3 |
+
size 1238564
|
referenceaudio/Finana/A4.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:29c41d404d2a3bf41fb4cb138d6a615a1b4fbad148f553201f85ddcf1b35bbd2
|
3 |
+
size 1497870
|
referenceaudio/Finana/A5.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:de7f216db700339df8cd93bbb2dc78c16e334edaa25597440bcfbb0af8da80ca
|
3 |
+
size 1663686
|
referenceaudio/Pippa/A1.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:908d33068d9fc9f32827b30b3a7170a1c5a41faff5b0ac8aa65d90233f0c9d33
|
3 |
+
size 1768452
|
referenceaudio/Pippa/A2.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b7c0ba14812c73c5e334b6dab30dd68f29f8ddd2517773353f6c3c1f886db3ed
|
3 |
+
size 1668612
|
referenceaudio/Pippa/A3.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c429758d5c304443c9897f6d15bba6c21159046b90835c387707abcf639b033d
|
3 |
+
size 1879806
|
referenceaudio/Pippa/A4.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8cf5030cb5a3fc612ffc60221ca6c0c023831e5bbbd022c46358e40de5fe75ca
|
3 |
+
size 1900926
|
referenceaudio/Pippa/A5.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cfdefdc76a973d10a0365e89bb17e0430927518c97ec874f1848b1198fe63ccc
|
3 |
+
size 1776124
|
voicelist.json
CHANGED
@@ -1,140 +1,184 @@
|
|
1 |
-
{
|
2 |
-
"Calli": {
|
3 |
-
"enable": true,
|
4 |
-
"gpt_model_path": "MoriCalliope_GPT.ckpt",
|
5 |
-
"sovits_model_path": "MoriCalliope_SoVITS.pth",
|
6 |
-
"title": "Mori Calliope",
|
7 |
-
"modelauthor": "Kit Lemonfoot",
|
8 |
-
"styles":{
|
9 |
-
"Neutral": "A1.wav",
|
10 |
-
"Soft": "A2.wav",
|
11 |
-
"Humored": "A3.wav",
|
12 |
-
"Preset 4": "A4.wav",
|
13 |
-
"Preset 5": "A5.wav",
|
14 |
-
"Preset 6": "A6.wav"
|
15 |
-
},
|
16 |
-
"styletrans":{
|
17 |
-
"Neutral": "But I commissioned it long before like, I commissioned it a long time before we released it, so like, I was right in the middle of my voice lessons.",
|
18 |
-
"Soft": "Sorry if the cicadas get to be loud or anything like that they're just really screeching.",
|
19 |
-
"Humored": "If you're sipping every time Mori spins un alive will kill you. Ruh roh raggy, Yeah I really did get down with the spins on that one huh?",
|
20 |
-
"Preset 4": "Um it's not, much of a spoiler so I wanna explain it now so that we can enjoy the song but the first part of the song, it starts out with the sound of a car right?",
|
21 |
-
"Preset 5": "Oh maybe we should do something besides rapping on this stage. Let's also sing.",
|
22 |
-
"Preset 6": "Yeah. But yeah I couldn't sing the highest note when I initially got the draft but I was like you know what no let's keep it."
|
23 |
-
},
|
24 |
-
"cover": "calli.png"
|
25 |
-
},
|
26 |
-
"Shiori": {
|
27 |
-
"enable": true,
|
28 |
-
"gpt_model_path": "ShioriNovella_GPT.ckpt",
|
29 |
-
"sovits_model_path": "ShioriNovella_SoVITS.pth",
|
30 |
-
"title": "Shiori Novella",
|
31 |
-
"modelauthor": "Kit Lemonfoot",
|
32 |
-
"styles":{
|
33 |
-
"Neutral": "A1.wav",
|
34 |
-
"Shocked": "A2.wav",
|
35 |
-
"Confused": "A3.wav",
|
36 |
-
"Scared": "A4.wav",
|
37 |
-
"Humored": "A5.wav",
|
38 |
-
"Thinking": "A6.wav"
|
39 |
-
},
|
40 |
-
"styletrans":{
|
41 |
-
"Neutral": "PlayStation was better Yeah that's true at the time. Ar Yong thank you for the S C Revelant And Christian thank you for becoming a bookworm.",
|
42 |
-
"Shocked": "I didn't know I forgot I didn't realize! I didn't realize! that's more comedic though!",
|
43 |
-
"Confused": "Isn't it banned in some places Oh I'm not sure. I'm not really- I really don't know. Why would it be banned? Because it has your imagination?",
|
44 |
-
"Scared": "Oh no! Oh no! There's another scene! Oh! Oh! Oh no Hide your children and hide your, your people and stuff! Oh sorry!",
|
45 |
-
"Humored": "Night Raven, Blitz Phantom, Rubber ton, thank you for the thing! Oh oh it starts off as an anime!",
|
46 |
-
"Thinking": "I can't... It seems like there's a slope there that she could climb out of right? Oh unless he, closes it."
|
47 |
-
},
|
48 |
-
"cover": "shiori.png"
|
49 |
-
},
|
50 |
-
"Zeta": {
|
51 |
-
"enable": true,
|
52 |
-
"gpt_model_path": "VestiaZeta_GPT.ckpt",
|
53 |
-
"sovits_model_path": "VestiaZeta_SoVITS.pth",
|
54 |
-
"title": "Vestia Zeta",
|
55 |
-
"modelauthor": "Kit Lemonfoot",
|
56 |
-
"styles":{
|
57 |
-
"Neutral": "A1.wav",
|
58 |
-
"Soft": "A2.wav",
|
59 |
-
"Confused": "A3.wav",
|
60 |
-
"Sarcastic": "A4.wav",
|
61 |
-
"Proud": "A5.wav"
|
62 |
-
},
|
63 |
-
"styletrans":{
|
64 |
-
"Neutral": "I think it's a good idea to have the real there because sometimes in movies sometimes they don't reload.",
|
65 |
-
"Soft": "No B G M just silence just me and my voice alone.",
|
66 |
-
"Confused": "Hunt the thimble what does that even mean, I mean I don't know English speakers help me out here what's hunt, the thimble?",
|
67 |
-
"Sarcastic": "Surely they're not gonna be shooting in this train. It's packed with people surely they won't be shooting here.",
|
68 |
-
"Proud": "Hell yeah we'll keep doing what we always do! Yeah boys!"
|
69 |
-
},
|
70 |
-
"cover": "zeta.png"
|
71 |
-
},
|
72 |
-
"
|
73 |
-
"enable": true,
|
74 |
-
"gpt_model_path": "
|
75 |
-
"sovits_model_path": "
|
76 |
-
"title": "
|
77 |
-
"modelauthor": "Kit Lemonfoot",
|
78 |
-
"styles":{
|
79 |
-
"Neutral": "A1.wav",
|
80 |
-
"
|
81 |
-
"
|
82 |
-
"
|
83 |
-
"Preset 5": "A5.wav"
|
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 |
-
"cover": "
|
139 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
140 |
}
|
|
|
1 |
+
{
|
2 |
+
"Calli": {
|
3 |
+
"enable": true,
|
4 |
+
"gpt_model_path": "MoriCalliope_GPT.ckpt",
|
5 |
+
"sovits_model_path": "MoriCalliope_SoVITS.pth",
|
6 |
+
"title": "Mori Calliope",
|
7 |
+
"modelauthor": "Kit Lemonfoot",
|
8 |
+
"styles":{
|
9 |
+
"Neutral": "A1.wav",
|
10 |
+
"Soft": "A2.wav",
|
11 |
+
"Humored": "A3.wav",
|
12 |
+
"Preset 4": "A4.wav",
|
13 |
+
"Preset 5": "A5.wav",
|
14 |
+
"Preset 6": "A6.wav"
|
15 |
+
},
|
16 |
+
"styletrans":{
|
17 |
+
"Neutral": "But I commissioned it long before like, I commissioned it a long time before we released it, so like, I was right in the middle of my voice lessons.",
|
18 |
+
"Soft": "Sorry if the cicadas get to be loud or anything like that they're just really screeching.",
|
19 |
+
"Humored": "If you're sipping every time Mori spins un alive will kill you. Ruh roh raggy, Yeah I really did get down with the spins on that one huh?",
|
20 |
+
"Preset 4": "Um it's not, much of a spoiler so I wanna explain it now so that we can enjoy the song but the first part of the song, it starts out with the sound of a car right?",
|
21 |
+
"Preset 5": "Oh maybe we should do something besides rapping on this stage. Let's also sing.",
|
22 |
+
"Preset 6": "Yeah. But yeah I couldn't sing the highest note when I initially got the draft but I was like you know what no let's keep it."
|
23 |
+
},
|
24 |
+
"cover": "calli.png"
|
25 |
+
},
|
26 |
+
"Shiori": {
|
27 |
+
"enable": true,
|
28 |
+
"gpt_model_path": "ShioriNovella_GPT.ckpt",
|
29 |
+
"sovits_model_path": "ShioriNovella_SoVITS.pth",
|
30 |
+
"title": "Shiori Novella",
|
31 |
+
"modelauthor": "Kit Lemonfoot",
|
32 |
+
"styles":{
|
33 |
+
"Neutral": "A1.wav",
|
34 |
+
"Shocked": "A2.wav",
|
35 |
+
"Confused": "A3.wav",
|
36 |
+
"Scared": "A4.wav",
|
37 |
+
"Humored": "A5.wav",
|
38 |
+
"Thinking": "A6.wav"
|
39 |
+
},
|
40 |
+
"styletrans":{
|
41 |
+
"Neutral": "PlayStation was better Yeah that's true at the time. Ar Yong thank you for the S C Revelant And Christian thank you for becoming a bookworm.",
|
42 |
+
"Shocked": "I didn't know I forgot I didn't realize! I didn't realize! that's more comedic though!",
|
43 |
+
"Confused": "Isn't it banned in some places Oh I'm not sure. I'm not really- I really don't know. Why would it be banned? Because it has your imagination?",
|
44 |
+
"Scared": "Oh no! Oh no! There's another scene! Oh! Oh! Oh no Hide your children and hide your, your people and stuff! Oh sorry!",
|
45 |
+
"Humored": "Night Raven, Blitz Phantom, Rubber ton, thank you for the thing! Oh oh it starts off as an anime!",
|
46 |
+
"Thinking": "I can't... It seems like there's a slope there that she could climb out of right? Oh unless he, closes it."
|
47 |
+
},
|
48 |
+
"cover": "shiori.png"
|
49 |
+
},
|
50 |
+
"Zeta": {
|
51 |
+
"enable": true,
|
52 |
+
"gpt_model_path": "VestiaZeta_GPT.ckpt",
|
53 |
+
"sovits_model_path": "VestiaZeta_SoVITS.pth",
|
54 |
+
"title": "Vestia Zeta",
|
55 |
+
"modelauthor": "Kit Lemonfoot",
|
56 |
+
"styles":{
|
57 |
+
"Neutral": "A1.wav",
|
58 |
+
"Soft": "A2.wav",
|
59 |
+
"Confused": "A3.wav",
|
60 |
+
"Sarcastic": "A4.wav",
|
61 |
+
"Proud": "A5.wav"
|
62 |
+
},
|
63 |
+
"styletrans":{
|
64 |
+
"Neutral": "I think it's a good idea to have the real there because sometimes in movies sometimes they don't reload.",
|
65 |
+
"Soft": "No B G M just silence just me and my voice alone.",
|
66 |
+
"Confused": "Hunt the thimble what does that even mean, I mean I don't know English speakers help me out here what's hunt, the thimble?",
|
67 |
+
"Sarcastic": "Surely they're not gonna be shooting in this train. It's packed with people surely they won't be shooting here.",
|
68 |
+
"Proud": "Hell yeah we'll keep doing what we always do! Yeah boys!"
|
69 |
+
},
|
70 |
+
"cover": "zeta.png"
|
71 |
+
},
|
72 |
+
"Finana": {
|
73 |
+
"enable": true,
|
74 |
+
"gpt_model_path": "FinanaRyugu_GPT.ckpt",
|
75 |
+
"sovits_model_path": "FinanaRyugu_SoVITS.pth",
|
76 |
+
"title": "Finana Ryugu",
|
77 |
+
"modelauthor": "Kit Lemonfoot",
|
78 |
+
"styles":{
|
79 |
+
"Neutral": "A1.wav",
|
80 |
+
"Soft": "A2.wav",
|
81 |
+
"Humored": "A3.wav",
|
82 |
+
"Preset 4": "A4.wav",
|
83 |
+
"Preset 5": "A5.wav"
|
84 |
+
},
|
85 |
+
"styletrans":{
|
86 |
+
"Neutral": "To catch up though, if you want to catch up on the stream delay I think you just have to set the playback speed to two times.",
|
87 |
+
"Soft": "Um, I'm really glad we get to do a watchalong today because my arm hurts, and my body aches so I'm kind of glad I'm not doing anything intense today.",
|
88 |
+
"Humored": "It's literally just a slideshow of a faded transition from each name!",
|
89 |
+
"Preset 4": "Wait, do I sing in, Hold on. Do I sing Happy Birthday in Finance too? I guess so just to, lead up to the B2 off okay?",
|
90 |
+
"Preset 5": "Anyway if you, wanna watch this with the Ryu Guard homies I'm pretty sure, some of you guys might be, in a group call right now watching this together so."
|
91 |
+
},
|
92 |
+
"cover": "finana.png"
|
93 |
+
},
|
94 |
+
"Pippa": {
|
95 |
+
"enable": true,
|
96 |
+
"gpt_model_path": "PipkinPippa_GPT.ckpt",
|
97 |
+
"sovits_model_path": "PipkinPippa_SoVITS.pth",
|
98 |
+
"title": "Pipkin Pippa",
|
99 |
+
"modelauthor": "Kit Lemonfoot",
|
100 |
+
"styles":{
|
101 |
+
"Neutral": "A1.wav",
|
102 |
+
"Soft": "A2.wav",
|
103 |
+
"Elevated": "A3.wav",
|
104 |
+
"Preset 4": "A4.wav",
|
105 |
+
"Preset 5": "A5.wav"
|
106 |
+
},
|
107 |
+
"styletrans":{
|
108 |
+
"Neutral": "He keeps coming over and he's doing like surprise visits and like, the children are home alone and the psychiatrist comes over and like the parents come back and they're like what the fuck are you doing in our house?",
|
109 |
+
"Soft": "Future as a script writer is as bright as your, perspecitve comedy, Forces I'm not trying to be a fucking script writer I'm just trying to talk to you guys about like the movie.",
|
110 |
+
"Elevated": "All he wanted to do was find his nieces he never gave up hope even though they were probably dead spent five years searching for them foundwith his crazy fucking bitch in law.",
|
111 |
+
"Preset 4": "Like I don't, I don't even have anybody to get drugs from even if I, even if I like did pursue that, right? I don't, I don't, I'm not, I'm not in Australia so I can't just ask Cleo.",
|
112 |
+
"Preset 5": "And, just like the build up to it and everything had her fucking laughing her ass off, and she started like swerving She was laughing so hard and I'll never forget."
|
113 |
+
},
|
114 |
+
"cover": "pippa.png"
|
115 |
+
},
|
116 |
+
"Lia": {
|
117 |
+
"enable": true,
|
118 |
+
"gpt_model_path": "AsheliaRinkou_GPT.ckpt",
|
119 |
+
"sovits_model_path": "AsheliaRinkou_SoVITS.pth",
|
120 |
+
"title": "Ashelia Rinkou",
|
121 |
+
"modelauthor": "Kit Lemonfoot",
|
122 |
+
"styles":{
|
123 |
+
"Neutral": "A1.wav",
|
124 |
+
"Angry": "A2.wav",
|
125 |
+
"Questioning": "A3.wav",
|
126 |
+
"Shouting": "A4.wav",
|
127 |
+
"Preset 5": "A5.wav",
|
128 |
+
"Preset 6": "A6.wav"
|
129 |
+
},
|
130 |
+
"styletrans":{
|
131 |
+
"Neutral": "What the fuck do you mean mid. I I don't know. I don't know Lord of the Rings just, wasn't.",
|
132 |
+
"Angry": "It's absolutely not what I fucking meant You guys are absolutely fucking disgusting I hate you all.",
|
133 |
+
"Questioning": "Mascot right? Like Shy Lily has her super cute little shrimpy, right? You get the tiny little shrimpy tattooed on you, No one's ever gonna know.",
|
134 |
+
"Shouting": "Maybe if I posted on Twitter and at you No I would not do that Cologne chan thank you for four months No I would no do that Oh my god, stop it!",
|
135 |
+
"Preset 5": "No instead of an engagement ring I, really really really just, think I would prefer to have a womb tattoo.",
|
136 |
+
"Preset 6": "Cannot, cannot say but I assure you, I assure you if you lived on the west coast anywhere on the west coast, You learned about the gold rush."
|
137 |
+
},
|
138 |
+
"cover": "lia.png"
|
139 |
+
},
|
140 |
+
"Dokibird": {
|
141 |
+
"enable": true,
|
142 |
+
"gpt_model_path": "Dokibird_GPT.ckpt",
|
143 |
+
"sovits_model_path": "Dokibird_SoVITS.pth",
|
144 |
+
"title": "Dokibird",
|
145 |
+
"modelauthor": "dacoolkid44, Kit Lemonfoot",
|
146 |
+
"styles":{
|
147 |
+
"Neutral": "A1.wav",
|
148 |
+
"Humored": "A2.wav",
|
149 |
+
"Laughing": "A3.wav",
|
150 |
+
"Cry Laughing": "A4.wav",
|
151 |
+
"Tired": "A5.wav",
|
152 |
+
"Bored": "A6.wav",
|
153 |
+
"Surprised": "A7.wav"
|
154 |
+
},
|
155 |
+
"styletrans":{
|
156 |
+
"Neutral": "Yeah it does look very risky I'd rather just like turn fast because it feels like, they can just shoot you better.",
|
157 |
+
"Humored": "Is that, I did it guys! Welcome to the ember ship welcome welcome!",
|
158 |
+
"Laughing": "Unforgiving lands of Alaska, sorry Alaska ha ha ha! What the fuck is what the fuck?",
|
159 |
+
"Cry Laughing": "Ha ha sorry, oh shit that's just so funny, the care up seeds.",
|
160 |
+
"Tired": "But I just I couldn't, I couldn't, be I was not I couldn't bring myself into finishing it.",
|
161 |
+
"Bored": "Alright let me go look through, let me go look through the hashtag guys I didn't, I looked through a little bit but you know.",
|
162 |
+
"Surprised": "What the heck a fish sword? Dude my weapon has everything it can be a watering can and a fishing pole?"
|
163 |
+
},
|
164 |
+
"cover": "dokibird.png"
|
165 |
+
},
|
166 |
+
"Template": {
|
167 |
+
"enable": false,
|
168 |
+
"gpt_model_path": "model.ckpt",
|
169 |
+
"sovits_model_path": "model.pth",
|
170 |
+
"title": "Full Name",
|
171 |
+
"modelauthor": "Kit Lemonfoot",
|
172 |
+
"styles":{
|
173 |
+
"name": "audio.wav",
|
174 |
+
"name": "audio.wav",
|
175 |
+
"name": "audio.wav"
|
176 |
+
},
|
177 |
+
"styletrans":{
|
178 |
+
"name": "",
|
179 |
+
"name": "",
|
180 |
+
"name": ""
|
181 |
+
},
|
182 |
+
"cover": "image.png"
|
183 |
+
}
|
184 |
}
|