File size: 6,729 Bytes
8cb3d54
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import torch
import commons
import models

import math
from torch import nn
from torch.nn import functional as F

import modules
import attentions

from torch.nn import Conv1d, ConvTranspose1d, Conv2d
from torch.nn.utils import weight_norm, remove_weight_norm, spectral_norm
from commons import init_weights, get_padding


class TextEncoder(nn.Module):
  def __init__(self,
               n_vocab,
               out_channels,
               hidden_channels,
               filter_channels,
               n_heads,
               n_layers,
               kernel_size,
               p_dropout,
               emotion_embedding):
    super().__init__()
    self.n_vocab = n_vocab
    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.emotion_embedding = emotion_embedding

    if self.n_vocab != 0:
      self.emb = nn.Embedding(n_vocab, hidden_channels)
      if emotion_embedding:
        self.emo_proj = nn.Linear(1024, hidden_channels)
      nn.init.normal_(self.emb.weight, 0.0, hidden_channels ** -0.5)

    self.encoder = attentions.Encoder(
      hidden_channels,
      filter_channels,
      n_heads,
      n_layers,
      kernel_size,
      p_dropout)
    self.proj = nn.Conv1d(hidden_channels, out_channels * 2, 1)

  def forward(self, x, x_lengths, emotion_embedding=None):
    if self.n_vocab != 0:
      x = self.emb(x) * math.sqrt(self.hidden_channels)  # [b, t, h]
    if emotion_embedding is not None:
      print("emotion added")
      x = x + self.emo_proj(emotion_embedding.unsqueeze(1))
    x = torch.transpose(x, 1, -1)  # [b, h, t]
    x_mask = torch.unsqueeze(commons.sequence_mask(x_lengths, x.size(2)), 1).to(x.dtype)

    x = self.encoder(x * x_mask, x_mask)
    stats = self.proj(x) * x_mask

    m, logs = torch.split(stats, self.out_channels, dim=1)
    return x, m, logs, x_mask


class PosteriorEncoder(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):
    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 SynthesizerTrn(models.SynthesizerTrn):
  """
  Synthesizer for Training
  """

  def __init__(self,
               n_vocab,
               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,
               n_speakers=0,
               gin_channels=0,
               use_sdp=True,
               emotion_embedding=False,
               ONNX_dir="./ONNX_net/",
               **kwargs):

    super().__init__(
      n_vocab,
      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,
      n_speakers=n_speakers,
      gin_channels=gin_channels,
      use_sdp=use_sdp,
      **kwargs
    )
    self.ONNX_dir = ONNX_dir
    self.enc_p = TextEncoder(n_vocab,
                             inter_channels,
                             hidden_channels,
                             filter_channels,
                             n_heads,
                             n_layers,
                             kernel_size,
                             p_dropout,
                             emotion_embedding)
    self.enc_q = PosteriorEncoder(spec_channels, inter_channels, hidden_channels, 5, 1, 16, gin_channels=gin_channels)

  def infer(self, x, x_lengths, sid=None, noise_scale=1, length_scale=1, noise_scale_w=1., max_len=None,
            emotion_embedding=None):
    from ONNXVITS_utils import runonnx
    with torch.no_grad():
      x, m_p, logs_p, x_mask = self.enc_p(x, x_lengths, emotion_embedding)

    if self.n_speakers > 0:
      g = self.emb_g(sid).unsqueeze(-1)  # [b, h, 1]
    else:
      g = None

    # logw = self.dp(x, x_mask, g=g, reverse=True, noise_scale=noise_scale_w)
    logw = runonnx(f"{self.ONNX_dir}dp.onnx", x=x.numpy(), x_mask=x_mask.numpy(), g=g.numpy())
    logw = torch.from_numpy(logw[0])

    w = torch.exp(logw) * x_mask * length_scale
    w_ceil = torch.ceil(w)
    y_lengths = torch.clamp_min(torch.sum(w_ceil, [1, 2]), 1).long()
    y_mask = torch.unsqueeze(commons.sequence_mask(y_lengths, None), 1).to(x_mask.dtype)
    attn_mask = torch.unsqueeze(x_mask, 2) * torch.unsqueeze(y_mask, -1)
    attn = commons.generate_path(w_ceil, attn_mask)

    m_p = torch.matmul(attn.squeeze(1), m_p.transpose(1, 2)).transpose(1, 2)  # [b, t', t], [b, t, d] -> [b, d, t']
    logs_p = torch.matmul(attn.squeeze(1), logs_p.transpose(1, 2)).transpose(1,
                                                                             2)  # [b, t', t], [b, t, d] -> [b, d, t']

    z_p = m_p + torch.randn_like(m_p) * torch.exp(logs_p) * noise_scale

    # z = self.flow(z_p, y_mask, g=g, reverse=True)
    z = runonnx(f"{self.ONNX_dir}flow.onnx", z_p=z_p.numpy(), y_mask=y_mask.numpy(), g=g.numpy())
    z = torch.from_numpy(z[0])

    # o = self.dec((z * y_mask)[:,:,:max_len], g=g)
    o = runonnx(f"{self.ONNX_dir}dec.onnx", z_in=(z * y_mask)[:, :, :max_len].numpy(), g=g.numpy())
    o = torch.from_numpy(o[0])

    return o, attn, y_mask, (z, z_p, m_p, logs_p)