fighter-programmer/voicegen
0
1import math2import torch3from torch import nn4from torch.nn import functional as F5 6from torch.nn import Conv1d7from torch.nn.utils import weight_norm, remove_weight_norm8 9import commons10from commons import init_weights, get_padding11from transforms import piecewise_rational_quadratic_transform12 13 14LRELU_SLOPE = 0.115 16 17class LayerNorm(nn.Module):18 def __init__(self, channels, eps=1e-5):19 super().__init__()20 self.channels = channels21 self.eps = eps22 23 self.gamma = nn.Parameter(torch.ones(channels))24 self.beta = nn.Parameter(torch.zeros(channels))25 26 def forward(self, x):27 x = x.transpose(1, -1)28 x = F.layer_norm(x, (self.channels,), self.gamma, self.beta, self.eps)29 return x.transpose(1, -1)30 31 32class ConvReluNorm(nn.Module):33 def __init__(self, in_channels, hidden_channels, out_channels, kernel_size, n_layers, p_dropout):34 super().__init__()35 self.in_channels = in_channels36 self.hidden_channels = hidden_channels37 self.out_channels = out_channels38 self.kernel_size = kernel_size39 self.n_layers = n_layers40 self.p_dropout = p_dropout41 assert n_layers > 1, "Number of layers should be larger than 0."42 43 self.conv_layers = nn.ModuleList()44 self.norm_layers = nn.ModuleList()45 self.conv_layers.append(nn.Conv1d(in_channels, hidden_channels, kernel_size, padding=kernel_size//2))46 self.norm_layers.append(LayerNorm(hidden_channels))47 self.relu_drop = nn.Sequential(48 nn.ReLU(),49 nn.Dropout(p_dropout))50 for _ in range(n_layers-1):51 self.conv_layers.append(nn.Conv1d(hidden_channels, hidden_channels, kernel_size, padding=kernel_size//2))52 self.norm_layers.append(LayerNorm(hidden_channels))53 self.proj = nn.Conv1d(hidden_channels, out_channels, 1)54 self.proj.weight.data.zero_()55 self.proj.bias.data.zero_()56 57 def forward(self, x, x_mask):58 x_org = x59 for i in range(self.n_layers):60 x = self.conv_layers[i](x * x_mask)61 x = self.norm_layers[i](x)62 x = self.relu_drop(x)63 x = x_org + self.proj(x)64 return x * x_mask65 66 67class DDSConv(nn.Module):68 """69 Dialted and Depth-Separable Convolution70 """71 def __init__(self, channels, kernel_size, n_layers, p_dropout=0.):72 super().__init__()73 self.channels = channels74 self.kernel_size = kernel_size75 self.n_layers = n_layers76 self.p_dropout = p_dropout77 78 self.drop = nn.Dropout(p_dropout)79 self.convs_sep = nn.ModuleList()80 self.convs_1x1 = nn.ModuleList()81 self.norms_1 = nn.ModuleList()82 self.norms_2 = nn.ModuleList()83 for i in range(n_layers):84 dilation = kernel_size ** i85 padding = (kernel_size * dilation - dilation) // 286 self.convs_sep.append(nn.Conv1d(channels, channels, kernel_size, 87 groups=channels, dilation=dilation, padding=padding88 ))89 self.convs_1x1.append(nn.Conv1d(channels, channels, 1))90 self.norms_1.append(LayerNorm(channels))91 self.norms_2.append(LayerNorm(channels))92 93 def forward(self, x, x_mask, g=None):94 if g is not None:95 x = x + g96 for i in range(self.n_layers):97 y = self.convs_sep[i](x * x_mask)98 y = self.norms_1[i](y)99 y = F.gelu(y)100 y = self.convs_1x1[i](y)101 y = self.norms_2[i](y)102 y = F.gelu(y)103 y = self.drop(y)104 x = x + y105 return x * x_mask106 107 108class WN(torch.nn.Module):109 def __init__(self, hidden_channels, kernel_size, dilation_rate, n_layers, gin_channels=0, p_dropout=0):110 super(WN, self).__init__()111 assert(kernel_size % 2 == 1)112 self.hidden_channels =hidden_channels113 self.kernel_size = kernel_size,114 self.dilation_rate = dilation_rate115 self.n_layers = n_layers116 self.gin_channels = gin_channels117 self.p_dropout = p_dropout118 119 self.in_layers = torch.nn.ModuleList()120 self.res_skip_layers = torch.nn.ModuleList()121 self.drop = nn.Dropout(p_dropout)122 123 if gin_channels != 0:124 cond_layer = torch.nn.Conv1d(gin_channels, 2*hidden_channels*n_layers, 1)125 self.cond_layer = torch.nn.utils.weight_norm(cond_layer, name='weight')126 127 for i in range(n_layers):128 dilation = dilation_rate ** i129 padding = int((kernel_size * dilation - dilation) / 2)130 in_layer = torch.nn.Conv1d(hidden_channels, 2*hidden_channels, kernel_size,131 dilation=dilation, padding=padding)132 in_layer = torch.nn.utils.weight_norm(in_layer, name='weight')133 self.in_layers.append(in_layer)134 135 # last one is not necessary136 if i < n_layers - 1:137 res_skip_channels = 2 * hidden_channels138 else:139 res_skip_channels = hidden_channels140 141 res_skip_layer = torch.nn.Conv1d(hidden_channels, res_skip_channels, 1)142 res_skip_layer = torch.nn.utils.weight_norm(res_skip_layer, name='weight')143 self.res_skip_layers.append(res_skip_layer)144 145 def forward(self, x, x_mask, g=None, **kwargs):146 output = torch.zeros_like(x)147 n_channels_tensor = torch.IntTensor([self.hidden_channels])148 149 if g is not None:150 g = self.cond_layer(g)151 152 for i in range(self.n_layers):153 x_in = self.in_layers[i](x)154 if g is not None:155 cond_offset = i * 2 * self.hidden_channels156 g_l = g[:,cond_offset:cond_offset+2*self.hidden_channels,:]157 else:158 g_l = torch.zeros_like(x_in)159 160 acts = commons.fused_add_tanh_sigmoid_multiply(161 x_in,162 g_l,163 n_channels_tensor)164 acts = self.drop(acts)165 166 res_skip_acts = self.res_skip_layers[i](acts)167 if i < self.n_layers - 1:168 res_acts = res_skip_acts[:,:self.hidden_channels,:]169 x = (x + res_acts) * x_mask170 output = output + res_skip_acts[:,self.hidden_channels:,:]171 else:172 output = output + res_skip_acts173 return output * x_mask174 175 def remove_weight_norm(self):176 if self.gin_channels != 0:177 torch.nn.utils.remove_weight_norm(self.cond_layer)178 for l in self.in_layers:179 torch.nn.utils.remove_weight_norm(l)180 for l in self.res_skip_layers:181 torch.nn.utils.remove_weight_norm(l)182 183 184class ResBlock1(torch.nn.Module):185 def __init__(self, channels, kernel_size=3, dilation=(1, 3, 5)):186 super(ResBlock1, self).__init__()187 self.convs1 = nn.ModuleList([188 weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[0],189 padding=get_padding(kernel_size, dilation[0]))),190 weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[1],191 padding=get_padding(kernel_size, dilation[1]))),192 weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[2],193 padding=get_padding(kernel_size, dilation[2])))194 ])195 self.convs1.apply(init_weights)196 197 self.convs2 = nn.ModuleList([198 weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=1,199 padding=get_padding(kernel_size, 1))),200 weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=1,201 padding=get_padding(kernel_size, 1))),202 weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=1,203 padding=get_padding(kernel_size, 1)))204 ])205 self.convs2.apply(init_weights)206 207 def forward(self, x, x_mask=None):208 for c1, c2 in zip(self.convs1, self.convs2):209 xt = F.leaky_relu(x, LRELU_SLOPE)210 if x_mask is not None:211 xt = xt * x_mask212 xt = c1(xt)213 xt = F.leaky_relu(xt, LRELU_SLOPE)214 if x_mask is not None:215 xt = xt * x_mask216 xt = c2(xt)217 x = xt + x218 if x_mask is not None:219 x = x * x_mask220 return x221 222 def remove_weight_norm(self):223 for l in self.convs1:224 remove_weight_norm(l)225 for l in self.convs2:226 remove_weight_norm(l)227 228 229class ResBlock2(torch.nn.Module):230 def __init__(self, channels, kernel_size=3, dilation=(1, 3)):231 super(ResBlock2, self).__init__()232 self.convs = nn.ModuleList([233 weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[0],234 padding=get_padding(kernel_size, dilation[0]))),235 weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[1],236 padding=get_padding(kernel_size, dilation[1])))237 ])238 self.convs.apply(init_weights)239 240 def forward(self, x, x_mask=None):241 for c in self.convs:242 xt = F.leaky_relu(x, LRELU_SLOPE)243 if x_mask is not None:244 xt = xt * x_mask245 xt = c(xt)246 x = xt + x247 if x_mask is not None:248 x = x * x_mask249 return x250 251 def remove_weight_norm(self):252 for l in self.convs:253 remove_weight_norm(l)254 255 256class Log(nn.Module):257 def forward(self, x, x_mask, reverse=False, **kwargs):258 if not reverse:259 y = torch.log(torch.clamp_min(x, 1e-5)) * x_mask260 logdet = torch.sum(-y, [1, 2])261 return y, logdet262 else:263 x = torch.exp(x) * x_mask264 return x265 266 267class Flip(nn.Module):268 def forward(self, x, *args, reverse=False, **kwargs):269 x = torch.flip(x, [1])270 if not reverse:271 logdet = torch.zeros(x.size(0)).to(dtype=x.dtype, device=x.device)272 return x, logdet273 else:274 return x275 276 277class ElementwiseAffine(nn.Module):278 def __init__(self, channels):279 super().__init__()280 self.channels = channels281 self.m = nn.Parameter(torch.zeros(channels,1))282 self.logs = nn.Parameter(torch.zeros(channels,1))283 284 def forward(self, x, x_mask, reverse=False, **kwargs):285 if not reverse:286 y = self.m + torch.exp(self.logs) * x287 y = y * x_mask288 logdet = torch.sum(self.logs * x_mask, [1,2])289 return y, logdet290 else:291 x = (x - self.m) * torch.exp(-self.logs) * x_mask292 return x293 294 295class ResidualCouplingLayer(nn.Module):296 def __init__(self,297 channels,298 hidden_channels,299 kernel_size,300 dilation_rate,301 n_layers,302 p_dropout=0,303 gin_channels=0,304 mean_only=False):305 assert channels % 2 == 0, "channels should be divisible by 2"306 super().__init__()307 self.channels = channels308 self.hidden_channels = hidden_channels309 self.kernel_size = kernel_size310 self.dilation_rate = dilation_rate311 self.n_layers = n_layers312 self.half_channels = channels // 2313 self.mean_only = mean_only314 315 self.pre = nn.Conv1d(self.half_channels, hidden_channels, 1)316 self.enc = WN(hidden_channels, kernel_size, dilation_rate, n_layers, p_dropout=p_dropout, gin_channels=gin_channels)317 self.post = nn.Conv1d(hidden_channels, self.half_channels * (2 - mean_only), 1)318 self.post.weight.data.zero_()319 self.post.bias.data.zero_()320 321 def forward(self, x, x_mask, g=None, reverse=False):322 x0, x1 = torch.split(x, [self.half_channels]*2, 1)323 h = self.pre(x0) * x_mask324 h = self.enc(h, x_mask, g=g)325 stats = self.post(h) * x_mask326 if not self.mean_only:327 m, logs = torch.split(stats, [self.half_channels]*2, 1)328 else:329 m = stats330 logs = torch.zeros_like(m)331 332 if not reverse:333 x1 = m + x1 * torch.exp(logs) * x_mask334 x = torch.cat([x0, x1], 1)335 logdet = torch.sum(logs, [1,2])336 return x, logdet337 else:338 x1 = (x1 - m) * torch.exp(-logs) * x_mask339 x = torch.cat([x0, x1], 1)340 return x341 342 343class ConvFlow(nn.Module):344 def __init__(self, in_channels, filter_channels, kernel_size, n_layers, num_bins=10, tail_bound=5.0):345 super().__init__()346 self.in_channels = in_channels347 self.filter_channels = filter_channels348 self.kernel_size = kernel_size349 self.n_layers = n_layers350 self.num_bins = num_bins351 self.tail_bound = tail_bound352 self.half_channels = in_channels // 2353 354 self.pre = nn.Conv1d(self.half_channels, filter_channels, 1)355 self.convs = DDSConv(filter_channels, kernel_size, n_layers, p_dropout=0.)356 self.proj = nn.Conv1d(filter_channels, self.half_channels * (num_bins * 3 - 1), 1)357 self.proj.weight.data.zero_()358 self.proj.bias.data.zero_()359 360 def forward(self, x, x_mask, g=None, reverse=False):361 x0, x1 = torch.split(x, [self.half_channels]*2, 1)362 h = self.pre(x0)363 h = self.convs(h, x_mask, g=g)364 h = self.proj(h) * x_mask365 366 b, c, t = x0.shape367 h = h.reshape(b, c, -1, t).permute(0, 1, 3, 2) # [b, cx?, t] -> [b, c, t, ?]368 369 unnormalized_widths = h[..., :self.num_bins] / math.sqrt(self.filter_channels)370 unnormalized_heights = h[..., self.num_bins:2*self.num_bins] / math.sqrt(self.filter_channels)371 unnormalized_derivatives = h[..., 2 * self.num_bins:]372 373 x1, logabsdet = piecewise_rational_quadratic_transform(x1,374 unnormalized_widths,375 unnormalized_heights,376 unnormalized_derivatives,377 inverse=reverse,378 tails='linear',379 tail_bound=self.tail_bound380 )381 382 x = torch.cat([x0, x1], 1) * x_mask383 logdet = torch.sum(logabsdet * x_mask, [1,2])384 if not reverse:385 return x, logdet386 else:387 return x388