File size: 8,517 Bytes
7ff2ba3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
from typing import Optional, List, Tuple

import torch
from torch import nn
from torch.nn import Conv1d, ConvTranspose1d
from torch.nn import functional as F
from torch.nn.utils import remove_weight_norm, weight_norm

from .residuals import ResBlock1, ResBlock2, LRELU_SLOPE
from .utils import call_weight_data_normal_if_Conv


class Generator(torch.nn.Module):
    def __init__(
        self,
        initial_channel: int,
        resblock: str,
        resblock_kernel_sizes: List[int],
        resblock_dilation_sizes: List[List[int]],
        upsample_rates: List[int],
        upsample_initial_channel: int,
        upsample_kernel_sizes: List[int],
        gin_channels: int = 0,
    ):
        super(Generator, self).__init__()
        self.num_kernels = len(resblock_kernel_sizes)
        self.num_upsamples = len(upsample_rates)

        self.conv_pre = Conv1d(
            initial_channel, upsample_initial_channel, 7, 1, padding=3
        )

        self.ups = nn.ModuleList()
        for i, (u, k) in enumerate(zip(upsample_rates, upsample_kernel_sizes)):
            self.ups.append(
                weight_norm(
                    ConvTranspose1d(
                        upsample_initial_channel // (2**i),
                        upsample_initial_channel // (2 ** (i + 1)),
                        k,
                        u,
                        padding=(k - u) // 2,
                    )
                )
            )

        self.resblocks = nn.ModuleList()
        resblock_module = ResBlock1 if resblock == "1" else ResBlock2
        for i in range(len(self.ups)):
            ch = upsample_initial_channel // (2 ** (i + 1))
            for k, d in zip(resblock_kernel_sizes, resblock_dilation_sizes):
                self.resblocks.append(resblock_module(ch, k, d))

        self.conv_post = Conv1d(ch, 1, 7, 1, padding=3, bias=False)
        self.ups.apply(call_weight_data_normal_if_Conv)

        if gin_channels != 0:
            self.cond = nn.Conv1d(gin_channels, upsample_initial_channel, 1)

    def __call__(
        self,
        x: torch.Tensor,
        g: Optional[torch.Tensor] = None,
        n_res: Optional[int] = None,
    ) -> torch.Tensor:
        return super().__call__(x, g=g, n_res=n_res)

    def forward(
        self,
        x: torch.Tensor,
        g: Optional[torch.Tensor] = None,
        n_res: Optional[int] = None,
    ):
        if n_res is not None:
            n = int(n_res)
            if n != x.shape[-1]:
                x = F.interpolate(x, size=n, mode="linear")

        x = self.conv_pre(x)
        if g is not None:
            x = x + self.cond(g)

        for i in range(self.num_upsamples):
            x = F.leaky_relu(x, LRELU_SLOPE)
            x = self.ups[i](x)
            n = i * self.num_kernels
            xs = self.resblocks[n](x)
            for j in range(1, self.num_kernels):
                xs += self.resblocks[n + j](x)
            x = xs / self.num_kernels

        x = F.leaky_relu(x)
        x = self.conv_post(x)
        x = torch.tanh(x)

        return x

    def __prepare_scriptable__(self):
        for l in self.ups:
            for hook in l._forward_pre_hooks.values():
                # The hook we want to remove is an instance of WeightNorm class, so
                # normally we would do `if isinstance(...)` but this class is not accessible
                # because of shadowing, so we check the module name directly.
                # https://github.com/pytorch/pytorch/blob/be0ca00c5ce260eb5bcec3237357f7a30cc08983/torch/nn/utils/__init__.py#L3
                if (
                    hook.__module__ == "torch.nn.utils.weight_norm"
                    and hook.__class__.__name__ == "WeightNorm"
                ):
                    torch.nn.utils.remove_weight_norm(l)

        for l in self.resblocks:
            for hook in l._forward_pre_hooks.values():
                if (
                    hook.__module__ == "torch.nn.utils.weight_norm"
                    and hook.__class__.__name__ == "WeightNorm"
                ):
                    torch.nn.utils.remove_weight_norm(l)
        return self

    def remove_weight_norm(self):
        for l in self.ups:
            remove_weight_norm(l)
        for l in self.resblocks:
            l.remove_weight_norm()


class SineGenerator(torch.nn.Module):
    """Definition of sine generator
    SineGenerator(samp_rate, harmonic_num = 0,
            sine_amp = 0.1, noise_std = 0.003,
            voiced_threshold = 0,
            flag_for_pulse=False)
    samp_rate: sampling rate in Hz
    harmonic_num: number of harmonic overtones (default 0)
    sine_amp: amplitude of sine-wavefrom (default 0.1)
    noise_std: std of Gaussian noise (default 0.003)
    voiced_thoreshold: F0 threshold for U/V classification (default 0)
    flag_for_pulse: this SinGen is used inside PulseGen (default False)
    Note: when flag_for_pulse is True, the first time step of a voiced
        segment is always sin(torch.pi) or cos(0)
    """

    def __init__(
        self,
        samp_rate: int,
        harmonic_num: int = 0,
        sine_amp: float = 0.1,
        noise_std: float = 0.003,
        voiced_threshold: int = 0,
    ):
        super(SineGenerator, self).__init__()
        self.sine_amp = sine_amp
        self.noise_std = noise_std
        self.harmonic_num = harmonic_num
        self.dim = harmonic_num + 1
        self.sampling_rate = samp_rate
        self.voiced_threshold = voiced_threshold

    def __call__(
        self, f0: torch.Tensor, upp: int
    ) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
        return super().__call__(f0, upp)

    def forward(
        self, f0: torch.Tensor, upp: int
    ) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
        """sine_tensor, uv = forward(f0)
        input F0: tensor(batchsize=1, length, dim=1)
                  f0 for unvoiced steps should be 0
        output sine_tensor: tensor(batchsize=1, length, dim)
        output uv: tensor(batchsize=1, length, 1)
        """
        with torch.no_grad():
            f0 = f0[:, None].transpose(1, 2)
            f0_buf = torch.zeros(f0.shape[0], f0.shape[1], self.dim, device=f0.device)
            # fundamental component
            f0_buf[:, :, 0] = f0[:, :, 0]
            for idx in range(self.harmonic_num):
                f0_buf[:, :, idx + 1] = f0_buf[:, :, 0] * (
                    idx + 2
                )  # idx + 2: the (idx+1)-th overtone, (idx+2)-th harmonic
            rad_values = (
                f0_buf / self.sampling_rate
            ) % 1  ###%1意味着n_har的乘积无法后处理优化
            rand_ini = torch.rand(
                f0_buf.shape[0], f0_buf.shape[2], device=f0_buf.device
            )
            rand_ini[:, 0] = 0
            rad_values[:, 0, :] = rad_values[:, 0, :] + rand_ini
            tmp_over_one = torch.cumsum(
                rad_values, 1
            )  # % 1  #####%1意味着后面的cumsum无法再优化
            tmp_over_one *= upp
            tmp_over_one: torch.Tensor = F.interpolate(
                tmp_over_one.transpose(2, 1),
                scale_factor=float(upp),
                mode="linear",
                align_corners=True,
            ).transpose(2, 1)
            rad_values: torch.Tensor = F.interpolate(
                rad_values.transpose(2, 1), scale_factor=float(upp), mode="nearest"
            ).transpose(
                2, 1
            )  #######
            tmp_over_one %= 1
            tmp_over_one_idx = (tmp_over_one[:, 1:, :] - tmp_over_one[:, :-1, :]) < 0
            cumsum_shift = torch.zeros_like(rad_values)
            cumsum_shift[:, 1:, :] = tmp_over_one_idx * -1.0
            sine_waves = torch.sin(
                torch.cumsum(rad_values + cumsum_shift, dim=1) * 2 * torch.pi
            )
            sine_waves = sine_waves * self.sine_amp
            uv = self._f02uv(f0)
            uv: torch.Tensor = F.interpolate(
                uv.transpose(2, 1), scale_factor=float(upp), mode="nearest"
            ).transpose(2, 1)
            noise_amp = uv * self.noise_std + (1 - uv) * self.sine_amp / 3
            noise = noise_amp * torch.randn_like(sine_waves)
            sine_waves = sine_waves * uv + noise
        return sine_waves, uv, noise

    def _f02uv(self, f0):
        # generate uv signal
        uv = torch.ones_like(f0)
        uv = uv * (f0 > self.voiced_threshold)
        if uv.device.type == "privateuseone":  # for DirectML
            uv = uv.float()
        return uv