Spaces:
Running
Running
File size: 13,034 Bytes
c5a22bf |
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 |
import torch
from torch import nn
from torch.nn import functional as F
import modules.attentions as attentions
import modules.commons as commons
import modules.modules as modules
from torch.nn import Conv1d, ConvTranspose1d, AvgPool1d, Conv2d
from torch.nn.utils import weight_norm, remove_weight_norm, spectral_norm
import utils
from modules.commons import init_weights, get_padding
from vdecoder.hifigan.models import Generator
from utils import f0_to_coarse
class ResidualCouplingBlock(nn.Module):
def __init__(self,
channels,
hidden_channels,
kernel_size,
dilation_rate,
n_layers,
n_flows=4,
gin_channels=0):
super().__init__()
self.channels = channels
self.hidden_channels = hidden_channels
self.kernel_size = kernel_size
self.dilation_rate = dilation_rate
self.n_layers = n_layers
self.n_flows = n_flows
self.gin_channels = gin_channels
self.flows = nn.ModuleList()
for i in range(n_flows):
self.flows.append(
modules.ResidualCouplingLayer(channels, hidden_channels, kernel_size, dilation_rate, n_layers,
gin_channels=gin_channels, mean_only=True))
self.flows.append(modules.Flip())
def forward(self, x, x_mask, g=None, reverse=False):
if not reverse:
for flow in self.flows:
x, _ = flow(x, x_mask, g=g, reverse=reverse)
else:
for flow in reversed(self.flows):
x = flow(x, x_mask, g=g, reverse=reverse)
return x
class Encoder(nn.Module):
def __init__(self,
in_channels,
out_channels,
hidden_channels,
kernel_size,
dilation_rate,
n_layers,
gin_channels=0):
super().__init__()
self.in_channels = in_channels
self.out_channels = out_channels
self.hidden_channels = hidden_channels
self.kernel_size = kernel_size
self.dilation_rate = dilation_rate
self.n_layers = n_layers
self.gin_channels = gin_channels
self.pre = nn.Conv1d(in_channels, hidden_channels, 1)
self.enc = modules.WN(hidden_channels, kernel_size, dilation_rate, n_layers, gin_channels=gin_channels)
self.proj = nn.Conv1d(hidden_channels, out_channels * 2, 1)
def forward(self, x, x_lengths, g=None):
# print(x.shape,x_lengths.shape)
x_mask = torch.unsqueeze(commons.sequence_mask(x_lengths, x.size(2)), 1).to(x.dtype)
x = self.pre(x) * x_mask
x = self.enc(x, x_mask, g=g)
stats = self.proj(x) * x_mask
m, logs = torch.split(stats, self.out_channels, dim=1)
z = (m + torch.randn_like(m) * torch.exp(logs)) * x_mask
return z, m, logs, x_mask
class TextEncoder(nn.Module):
def __init__(self,
out_channels,
hidden_channels,
kernel_size,
n_layers,
gin_channels=0,
filter_channels=None,
n_heads=None,
p_dropout=None):
super().__init__()
self.out_channels = out_channels
self.hidden_channels = hidden_channels
self.kernel_size = kernel_size
self.n_layers = n_layers
self.gin_channels = gin_channels
self.proj = nn.Conv1d(hidden_channels, out_channels * 2, 1)
self.f0_emb = nn.Embedding(256, hidden_channels)
self.enc_ = attentions.Encoder(
hidden_channels,
filter_channels,
n_heads,
n_layers,
kernel_size,
p_dropout)
def forward(self, x, x_mask, f0=None, z=None):
x = x + self.f0_emb(f0).transpose(1, 2)
x = self.enc_(x * x_mask, x_mask)
stats = self.proj(x) * x_mask
m, logs = torch.split(stats, self.out_channels, dim=1)
z = (m + z * torch.exp(logs)) * x_mask
return z, m, logs, x_mask
class DiscriminatorP(torch.nn.Module):
def __init__(self, period, kernel_size=5, stride=3, use_spectral_norm=False):
super(DiscriminatorP, self).__init__()
self.period = period
self.use_spectral_norm = use_spectral_norm
norm_f = weight_norm if use_spectral_norm == False else spectral_norm
self.convs = nn.ModuleList([
norm_f(Conv2d(1, 32, (kernel_size, 1), (stride, 1), padding=(get_padding(kernel_size, 1), 0))),
norm_f(Conv2d(32, 128, (kernel_size, 1), (stride, 1), padding=(get_padding(kernel_size, 1), 0))),
norm_f(Conv2d(128, 512, (kernel_size, 1), (stride, 1), padding=(get_padding(kernel_size, 1), 0))),
norm_f(Conv2d(512, 1024, (kernel_size, 1), (stride, 1), padding=(get_padding(kernel_size, 1), 0))),
norm_f(Conv2d(1024, 1024, (kernel_size, 1), 1, padding=(get_padding(kernel_size, 1), 0))),
])
self.conv_post = norm_f(Conv2d(1024, 1, (3, 1), 1, padding=(1, 0)))
def forward(self, x):
fmap = []
# 1d to 2d
b, c, t = x.shape
if t % self.period != 0: # pad first
n_pad = self.period - (t % self.period)
x = F.pad(x, (0, n_pad), "reflect")
t = t + n_pad
x = x.view(b, c, t // self.period, self.period)
for l in self.convs:
x = l(x)
x = F.leaky_relu(x, modules.LRELU_SLOPE)
fmap.append(x)
x = self.conv_post(x)
fmap.append(x)
x = torch.flatten(x, 1, -1)
return x, fmap
class DiscriminatorS(torch.nn.Module):
def __init__(self, use_spectral_norm=False):
super(DiscriminatorS, self).__init__()
norm_f = weight_norm if use_spectral_norm == False else spectral_norm
self.convs = nn.ModuleList([
norm_f(Conv1d(1, 16, 15, 1, padding=7)),
norm_f(Conv1d(16, 64, 41, 4, groups=4, padding=20)),
norm_f(Conv1d(64, 256, 41, 4, groups=16, padding=20)),
norm_f(Conv1d(256, 1024, 41, 4, groups=64, padding=20)),
norm_f(Conv1d(1024, 1024, 41, 4, groups=256, padding=20)),
norm_f(Conv1d(1024, 1024, 5, 1, padding=2)),
])
self.conv_post = norm_f(Conv1d(1024, 1, 3, 1, padding=1))
def forward(self, x):
fmap = []
for l in self.convs:
x = l(x)
x = F.leaky_relu(x, modules.LRELU_SLOPE)
fmap.append(x)
x = self.conv_post(x)
fmap.append(x)
x = torch.flatten(x, 1, -1)
return x, fmap
class F0Decoder(nn.Module):
def __init__(self,
out_channels,
hidden_channels,
filter_channels,
n_heads,
n_layers,
kernel_size,
p_dropout,
spk_channels=0):
super().__init__()
self.out_channels = out_channels
self.hidden_channels = hidden_channels
self.filter_channels = filter_channels
self.n_heads = n_heads
self.n_layers = n_layers
self.kernel_size = kernel_size
self.p_dropout = p_dropout
self.spk_channels = spk_channels
self.prenet = nn.Conv1d(hidden_channels, hidden_channels, 3, padding=1)
self.decoder = attentions.FFT(
hidden_channels,
filter_channels,
n_heads,
n_layers,
kernel_size,
p_dropout)
self.proj = nn.Conv1d(hidden_channels, out_channels, 1)
self.f0_prenet = nn.Conv1d(1, hidden_channels, 3, padding=1)
self.cond = nn.Conv1d(spk_channels, hidden_channels, 1)
def forward(self, x, norm_f0, x_mask, spk_emb=None):
x = torch.detach(x)
if spk_emb is not None:
x = x + self.cond(spk_emb)
x += self.f0_prenet(norm_f0)
x = self.prenet(x) * x_mask
x = self.decoder(x * x_mask, x_mask)
x = self.proj(x) * x_mask
return x
class SynthesizerTrn(nn.Module):
"""
Synthesizer for Training
"""
def __init__(self,
spec_channels,
segment_size,
inter_channels,
hidden_channels,
filter_channels,
n_heads,
n_layers,
kernel_size,
p_dropout,
resblock,
resblock_kernel_sizes,
resblock_dilation_sizes,
upsample_rates,
upsample_initial_channel,
upsample_kernel_sizes,
gin_channels,
ssl_dim,
n_speakers,
sampling_rate=44100,
**kwargs):
super().__init__()
self.spec_channels = spec_channels
self.inter_channels = inter_channels
self.hidden_channels = hidden_channels
self.filter_channels = filter_channels
self.n_heads = n_heads
self.n_layers = n_layers
self.kernel_size = kernel_size
self.p_dropout = p_dropout
self.resblock = resblock
self.resblock_kernel_sizes = resblock_kernel_sizes
self.resblock_dilation_sizes = resblock_dilation_sizes
self.upsample_rates = upsample_rates
self.upsample_initial_channel = upsample_initial_channel
self.upsample_kernel_sizes = upsample_kernel_sizes
self.segment_size = segment_size
self.gin_channels = gin_channels
self.ssl_dim = ssl_dim
self.emb_g = nn.Embedding(n_speakers, gin_channels)
self.pre = nn.Conv1d(ssl_dim, hidden_channels, kernel_size=5, padding=2)
self.enc_p = TextEncoder(
inter_channels,
hidden_channels,
filter_channels=filter_channels,
n_heads=n_heads,
n_layers=n_layers,
kernel_size=kernel_size,
p_dropout=p_dropout
)
hps = {
"sampling_rate": sampling_rate,
"inter_channels": inter_channels,
"resblock": resblock,
"resblock_kernel_sizes": resblock_kernel_sizes,
"resblock_dilation_sizes": resblock_dilation_sizes,
"upsample_rates": upsample_rates,
"upsample_initial_channel": upsample_initial_channel,
"upsample_kernel_sizes": upsample_kernel_sizes,
"gin_channels": gin_channels,
}
self.dec = Generator(h=hps)
self.enc_q = Encoder(spec_channels, inter_channels, hidden_channels, 5, 1, 16, gin_channels=gin_channels)
self.flow = ResidualCouplingBlock(inter_channels, hidden_channels, 5, 1, 4, gin_channels=gin_channels)
self.f0_decoder = F0Decoder(
1,
hidden_channels,
filter_channels,
n_heads,
n_layers,
kernel_size,
p_dropout,
spk_channels=gin_channels
)
self.emb_uv = nn.Embedding(2, hidden_channels)
self.predict_f0 = False
self.speaker_map = []
self.export_mix = False
def export_chara_mix(self, n_speakers_mix):
self.speaker_map = torch.zeros((n_speakers_mix, 1, 1, self.gin_channels))
for i in range(n_speakers_mix):
self.speaker_map[i] = self.emb_g(torch.LongTensor([[i]]))
self.speaker_map = self.speaker_map.unsqueeze(0)
self.export_mix = True
def forward(self, c, f0, mel2ph, uv, noise=None, g=None, cluster_infer_ratio=0.1):
decoder_inp = F.pad(c, [0, 0, 1, 0])
mel2ph_ = mel2ph.unsqueeze(2).repeat([1, 1, c.shape[-1]])
c = torch.gather(decoder_inp, 1, mel2ph_).transpose(1, 2) # [B, T, H]
c_lengths = (torch.ones(c.size(0)) * c.size(-1)).to(c.device)
if self.export_mix: # [N, S] * [S, B, 1, H]
g = g.reshape((g.shape[0], g.shape[1], 1, 1, 1)) # [N, S, B, 1, 1]
g = g * self.speaker_map # [N, S, B, 1, H]
g = torch.sum(g, dim=1) # [N, 1, B, 1, H]
g = g.transpose(0, -1).transpose(0, -2).squeeze(0) # [B, H, N]
else:
g = g.unsqueeze(0)
g = self.emb_g(g).transpose(1, 2)
x_mask = torch.unsqueeze(commons.sequence_mask(c_lengths, c.size(2)), 1).to(c.dtype)
x = self.pre(c) * x_mask + self.emb_uv(uv.long()).transpose(1, 2)
if self.predict_f0:
lf0 = 2595. * torch.log10(1. + f0.unsqueeze(1) / 700.) / 500
norm_lf0 = utils.normalize_f0(lf0, x_mask, uv, random_scale=False)
pred_lf0 = self.f0_decoder(x, norm_lf0, x_mask, spk_emb=g)
f0 = (700 * (torch.pow(10, pred_lf0 * 500 / 2595) - 1)).squeeze(1)
z_p, m_p, logs_p, c_mask = self.enc_p(x, x_mask, f0=f0_to_coarse(f0), z=noise)
z = self.flow(z_p, c_mask, g=g, reverse=True)
o = self.dec(z * c_mask, g=g, f0=f0)
return o
|