Spaces:
Running
on
Zero
Running
on
Zero
File size: 13,168 Bytes
37ced70 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 |
import typing as tp
import torch
import torch.nn as nn
from torch.nn import Conv1d, ConvTranspose1d
from torch.nn.utils import weight_norm, remove_weight_norm
from fireredtts.modules.bigvgan.alias_free_torch import (
Activation1d as TorchActivation1d,
)
from fireredtts.modules.bigvgan.activations import Snake, SnakeBeta
def init_weights(m, mean=0.0, std=0.01):
classname = m.__class__.__name__
if classname.find("Conv") != -1:
m.weight.data.normal_(mean, std)
def get_padding(kernel_size, dilation=1):
return int((kernel_size * dilation - dilation) / 2)
class AMPBlock1(torch.nn.Module):
def __init__(
self,
channels,
kernel_size=3,
dilation=(1, 3, 5),
activation=None,
snake_logscale=True,
use_cuda_kernel=False,
):
super(AMPBlock1, self).__init__()
self.convs1 = nn.ModuleList(
[
weight_norm(
Conv1d(
channels,
channels,
kernel_size,
1,
dilation=dilation[0],
padding=get_padding(kernel_size, dilation[0]),
)
),
weight_norm(
Conv1d(
channels,
channels,
kernel_size,
1,
dilation=dilation[1],
padding=get_padding(kernel_size, dilation[1]),
)
),
weight_norm(
Conv1d(
channels,
channels,
kernel_size,
1,
dilation=dilation[2],
padding=get_padding(kernel_size, dilation[2]),
)
),
]
)
self.convs1.apply(init_weights)
self.convs2 = nn.ModuleList(
[
weight_norm(
Conv1d(
channels,
channels,
kernel_size,
1,
dilation=1,
padding=get_padding(kernel_size, 1),
)
),
weight_norm(
Conv1d(
channels,
channels,
kernel_size,
1,
dilation=1,
padding=get_padding(kernel_size, 1),
)
),
weight_norm(
Conv1d(
channels,
channels,
kernel_size,
1,
dilation=1,
padding=get_padding(kernel_size, 1),
)
),
]
)
self.convs2.apply(init_weights)
self.num_layers = len(self.convs1) + len(
self.convs2
) # total number of conv layers
# select which Activation1d, lazy-load cuda version to ensure backward compatibility
if use_cuda_kernel:
from modules.bigvgan.alias_free_cuda.activation1d import (
Activation1d as CudaActivation1d,
)
Activation1d = CudaActivation1d
else:
Activation1d = TorchActivation1d
if (
activation == "snake"
): # periodic nonlinearity with snake function and anti-aliasing
self.activations = nn.ModuleList(
[
Activation1d(
activation=Snake(channels, alpha_logscale=snake_logscale)
)
for _ in range(self.num_layers)
]
)
elif (
activation == "snakebeta"
): # periodic nonlinearity with snakebeta function and anti-aliasing
self.activations = nn.ModuleList(
[
Activation1d(
activation=SnakeBeta(channels, alpha_logscale=snake_logscale)
)
for _ in range(self.num_layers)
]
)
else:
raise NotImplementedError(
"activation incorrectly specified. check the config file and look for 'activation'."
)
def forward(self, x):
acts1, acts2 = self.activations[::2], self.activations[1::2]
for c1, c2, a1, a2 in zip(self.convs1, self.convs2, acts1, acts2):
xt = a1(x)
xt = c1(xt)
xt = a2(xt)
xt = c2(xt)
x = xt + x
return x
def remove_weight_norm(self):
for l in self.convs1:
remove_weight_norm(l)
for l in self.convs2:
remove_weight_norm(l)
class AMPBlock2(torch.nn.Module):
def __init__(
self,
channels,
kernel_size=3,
dilation=(1, 3),
activation=None,
snake_logscale=True,
use_cuda_kernel=False,
):
super(AMPBlock2, self).__init__()
self.convs = nn.ModuleList(
[
weight_norm(
Conv1d(
channels,
channels,
kernel_size,
1,
dilation=dilation[0],
padding=get_padding(kernel_size, dilation[0]),
)
),
weight_norm(
Conv1d(
channels,
channels,
kernel_size,
1,
dilation=dilation[1],
padding=get_padding(kernel_size, dilation[1]),
)
),
]
)
self.convs.apply(init_weights)
self.num_layers = len(self.convs) # total number of conv layers
# select which Activation1d, lazy-load cuda version to ensure backward compatibility
if use_cuda_kernel:
from modules.bigvgan.alias_free_cuda.activation1d import (
Activation1d as CudaActivation1d,
)
Activation1d = CudaActivation1d
else:
Activation1d = TorchActivation1d
if (
activation == "snake"
): # periodic nonlinearity with snake function and anti-aliasing
self.activations = nn.ModuleList(
[
Activation1d(
activation=Snake(channels, alpha_logscale=snake_logscale)
)
for _ in range(self.num_layers)
]
)
elif (
activation == "snakebeta"
): # periodic nonlinearity with snakebeta function and anti-aliasing
self.activations = nn.ModuleList(
[
Activation1d(
activation=SnakeBeta(channels, alpha_logscale=snake_logscale)
)
for _ in range(self.num_layers)
]
)
else:
raise NotImplementedError(
"activation incorrectly specified. check the config file and look for 'activation'."
)
def forward(self, x):
for c, a in zip(self.convs, self.activations):
xt = a(x)
xt = c(xt)
x = xt + x
return x
def remove_weight_norm(self):
for l in self.convs:
remove_weight_norm(l)
class BigVGAN(torch.nn.Module):
# this is our main BigVGAN model. Applies anti-aliased periodic activation for resblocks.
def __init__(
self,
num_mels: int,
upsample_initial_channel: int,
resblock_kernel_sizes: tp.List[int],
resblock_dilation_sizes: tp.List[tp.List[int]],
upsample_rates: tp.List[int],
upsample_kernel_sizes: tp.List[int],
resblock_type: str = "1",
snake_logscale: bool = True,
activation: str = "snakebeta",
use_tanh_at_final: bool = False,
use_bias_at_final: bool = False,
use_cuda_kernel: bool = False,
):
super(BigVGAN, self).__init__()
self.num_kernels = len(resblock_kernel_sizes)
self.num_upsamples = len(upsample_rates)
# pre conv
self.conv_pre = weight_norm(
Conv1d(num_mels, upsample_initial_channel, 7, 1, padding=3)
)
# define which AMPBlock to use. BigVGAN uses AMPBlock1 as default
resblock = AMPBlock1 if resblock_type == "1" else AMPBlock2
# transposed conv-based upsamplers. does not apply anti-aliasing
self.ups = nn.ModuleList()
for i, (u, k) in enumerate(zip(upsample_rates, upsample_kernel_sizes)):
self.ups.append(
nn.ModuleList(
[
weight_norm(
ConvTranspose1d(
upsample_initial_channel // (2**i),
upsample_initial_channel // (2 ** (i + 1)),
k,
u,
padding=(k - u) // 2,
)
)
]
)
)
# residual blocks using anti-aliased multi-periodicity composition modules (AMP)
self.resblocks = nn.ModuleList()
for i in range(len(self.ups)):
ch = upsample_initial_channel // (2 ** (i + 1))
for j, (k, d) in enumerate(
zip(resblock_kernel_sizes, resblock_dilation_sizes)
):
self.resblocks.append(
resblock(
ch,
k,
d,
activation=activation,
snake_logscale=snake_logscale,
use_cuda_kernel=use_cuda_kernel,
)
)
# select which Activation1d, lazy-load cuda version to ensure backward compatibility
if use_cuda_kernel:
from modules.bigvgan.alias_free_cuda.activation1d import (
Activation1d as CudaActivation1d,
)
Activation1d = CudaActivation1d
else:
Activation1d = TorchActivation1d
# post conv
if (
activation == "snake"
): # periodic nonlinearity with snake function and anti-aliasing
activation_post = Snake(ch, alpha_logscale=snake_logscale)
self.activation_post = Activation1d(activation=activation_post)
elif (
activation == "snakebeta"
): # periodic nonlinearity with snakebeta function and anti-aliasing
activation_post = SnakeBeta(ch, alpha_logscale=snake_logscale)
self.activation_post = Activation1d(activation=activation_post)
else:
raise NotImplementedError(
"activation incorrectly specified. check the config file and look for 'activation'."
)
# whether to use bias for the final conv_post. Defaults to True for backward compatibility
self.use_bias_at_final = use_bias_at_final
self.conv_post = weight_norm(
Conv1d(ch, 1, 7, 1, padding=3, bias=self.use_bias_at_final)
)
# weight initialization
for i in range(len(self.ups)):
self.ups[i].apply(init_weights)
self.conv_post.apply(init_weights)
# final tanh activation. Defaults to True for backward compatibility
self.use_tanh_at_final = use_tanh_at_final
def forward(self, x):
# pre conv
x = self.conv_pre(x)
for i in range(self.num_upsamples):
# upsampling
for i_up in range(len(self.ups[i])):
x = self.ups[i][i_up](x)
# AMP blocks
xs = None
for j in range(self.num_kernels):
if xs is None:
xs = self.resblocks[i * self.num_kernels + j](x)
else:
xs += self.resblocks[i * self.num_kernels + j](x)
x = xs / self.num_kernels
# post conv
x = self.activation_post(x)
x = self.conv_post(x)
# final tanh activation
if self.use_tanh_at_final:
x = torch.tanh(x)
else:
x = torch.clamp(x, min=-1.0, max=1.0) # bound the output to [-1, 1]
return x
def remove_weight_norm(self):
print("Removing weight norm...")
for l in self.ups:
for l_i in l:
remove_weight_norm(l_i)
for l in self.resblocks:
l.remove_weight_norm()
remove_weight_norm(self.conv_pre)
remove_weight_norm(self.conv_post)
|