from logging import getLogger
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.nn import Conv1d, ConvTranspose1d
from torch.nn.utils import remove_weight_norm, weight_norm
from ...modules import ResBlock1, ResBlock2
from ._utils import init_weights
LOG = getLogger(__name__)
LRELU_SLOPE = 0.1
def padDiff(x):
return F.pad(F.pad(x, (0, 0, -1, 1), "constant", 0) - x, (0, 0, 0, -1), "constant", 0)
class SineGen(torch.nn.Module):
"""
Definition of sine generator
SineGen(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(np.pi) or cos(0)
"""
def __init__(
self,
samp_rate,
harmonic_num=0,
sine_amp=0.1,
noise_std=0.003,
voiced_threshold=0,
flag_for_pulse=False,
):
super().__init__()
self.sine_amp = sine_amp
self.noise_std = noise_std
self.harmonic_num = harmonic_num
self.dim = self.harmonic_num + 1
self.sampling_rate = samp_rate
self.voiced_threshold = voiced_threshold
self.flag_for_pulse = flag_for_pulse
def _f02uv(self, f0):
# generate uv signal
uv = (f0 > self.voiced_threshold).type(torch.float32)
return uv
def _f02sine(self, f0_values):
"""
f0_values: (batchsize, length, dim)
where dim indicates fundamental tone and overtones
"""
# convert to F0 in rad. The integer part n can be ignored
# because 2 * np.pi * n doesn't affect phase
rad_values = (f0_values / self.sampling_rate) % 1
# initial phase noise (no noise for fundamental component)
rand_ini = torch.rand(f0_values.shape[0], f0_values.shape[2], device=f0_values.device)
rand_ini[:, 0] = 0
rad_values[:, 0, :] = rad_values[:, 0, :] + rand_ini
# instantanouse phase sine[t] = sin(2*pi \sum_i=1 ^{t} rad)
if not self.flag_for_pulse:
# for normal case
# To prevent torch.cumsum numerical overflow,
# it is necessary to add -1 whenever \sum_k=1^n rad_value_k > 1.
# Buffer tmp_over_one_idx indicates the time step to add -1.
# This will not change F0 of sine because (x-1) * 2*pi = x * 2*pi
tmp_over_one = torch.cumsum(rad_values, 1) % 1
tmp_over_one_idx = (padDiff(tmp_over_one)) < 0
cumsum_shift = torch.zeros_like(rad_values)
cumsum_shift[:, 1:, :] = tmp_over_one_idx * -1.0
sines = torch.sin(torch.cumsum(rad_values + cumsum_shift, dim=1) * 2 * np.pi)
else:
# If necessary, make sure that the first time step of every
# voiced segments is sin(pi) or cos(0)
# This is used for pulse-train generation
# identify the last time step in unvoiced segments
uv = self._f02uv(f0_values)
uv_1 = torch.roll(uv, shifts=-1, dims=1)
uv_1[:, -1, :] = 1
u_loc = (uv < 1) * (uv_1 > 0)
# get the instantanouse phase
tmp_cumsum = torch.cumsum(rad_values, dim=1)
# different batch needs to be processed differently
for idx in range(f0_values.shape[0]):
temp_sum = tmp_cumsum[idx, u_loc[idx, :, 0], :]
temp_sum[1:, :] = temp_sum[1:, :] - temp_sum[0:-1, :]
# stores the accumulation of i.phase within
# each voiced segments
tmp_cumsum[idx, :, :] = 0
tmp_cumsum[idx, u_loc[idx, :, 0], :] = temp_sum
# rad_values - tmp_cumsum: remove the accumulation of i.phase
# within the previous voiced segment.
i_phase = torch.cumsum(rad_values - tmp_cumsum, dim=1)
# get the sines
sines = torch.cos(i_phase * 2 * np.pi)
return sines
def forward(self, f0):
"""
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_buf = torch.zeros(f0.shape[0], f0.shape[1], self.dim, device=f0.device)
# fundamental component
# fn = torch.multiply(
# f0, torch.FloatTensor([[range(1, self.harmonic_num + 2)]]).to(f0.device)
# )
fn = torch.multiply(f0, torch.arange(1, self.harmonic_num + 2).to(f0.device).to(f0.dtype))
# generate sine waveforms
sine_waves = self._f02sine(fn) * self.sine_amp
# generate uv signal
# uv = torch.ones(f0.shape)
# uv = uv * (f0 > self.voiced_threshold)
uv = self._f02uv(f0)
# noise: for unvoiced should be similar to sine_amp
# std = self.sine_amp/3 -> max value ~ self.sine_amp
# . for voiced regions is self.noise_std
noise_amp = uv * self.noise_std + (1 - uv) * self.sine_amp / 3
noise = noise_amp * torch.randn_like(sine_waves)
# first: set the unvoiced part to 0 by uv
# then: additive noise
sine_waves = sine_waves * uv + noise
return sine_waves, uv, noise
class SourceModuleHnNSF(torch.nn.Module):
"""
SourceModule for hn-nsf
SourceModule(sampling_rate, harmonic_num=0, sine_amp=0.1,
add_noise_std=0.003, voiced_threshod=0)
sampling_rate: sampling_rate in Hz
harmonic_num: number of harmonic above F0 (default: 0)
sine_amp: amplitude of sine source signal (default: 0.1)
add_noise_std: std of additive Gaussian noise (default: 0.003)
note that amplitude of noise in unvoiced is decided
by sine_amp
voiced_threshold: threshold to set U/V given F0 (default: 0)
Sine_source, noise_source = SourceModuleHnNSF(F0_sampled)
F0_sampled (batchsize, length, 1)
Sine_source (batchsize, length, 1)
noise_source (batchsize, length 1)
uv (batchsize, length, 1)
"""
def __init__(
self,
sampling_rate,
harmonic_num=0,
sine_amp=0.1,
add_noise_std=0.003,
voiced_threshod=0,
):
super().__init__()
self.sine_amp = sine_amp
self.noise_std = add_noise_std
# to produce sine waveforms
self.l_sin_gen = SineGen(sampling_rate, harmonic_num, sine_amp, add_noise_std, voiced_threshod)
# to merge source harmonics into a single excitation
self.l_linear = torch.nn.Linear(harmonic_num + 1, 1)
self.l_tanh = torch.nn.Tanh()
def forward(self, x):
"""
Sine_source, noise_source = SourceModuleHnNSF(F0_sampled)
F0_sampled (batchsize, length, 1)
Sine_source (batchsize, length, 1)
noise_source (batchsize, length 1)
"""
# source for harmonic branch
sine_wavs, uv, _ = self.l_sin_gen(x)
sine_merge = self.l_tanh(self.l_linear(sine_wavs))
# source for noise branch, in the same shape as uv
noise = torch.randn_like(uv) * self.sine_amp / 3
return sine_merge, noise, uv
[docs]
class NSFHifiGANGenerator(torch.nn.Module):
def __init__(self, h):
super().__init__()
self.h = h
self.num_kernels = len(h["resblock_kernel_sizes"])
self.num_upsamples = len(h["upsample_rates"])
self.f0_upsamp = torch.nn.Upsample(scale_factor=np.prod(h["upsample_rates"]))
self.m_source = SourceModuleHnNSF(sampling_rate=h["sampling_rate"], harmonic_num=8)
self.noise_convs = nn.ModuleList()
self.conv_pre = weight_norm(Conv1d(h["inter_channels"], h["upsample_initial_channel"], 7, 1, padding=3))
resblock = ResBlock1 if h["resblock"] == "1" else ResBlock2
self.ups = nn.ModuleList()
for i, (u, k) in enumerate(zip(h["upsample_rates"], h["upsample_kernel_sizes"])):
c_cur = h["upsample_initial_channel"] // (2 ** (i + 1))
self.ups.append(
weight_norm(
ConvTranspose1d(
h["upsample_initial_channel"] // (2**i),
h["upsample_initial_channel"] // (2 ** (i + 1)),
k,
u,
padding=(k - u) // 2,
)
)
)
if i + 1 < len(h["upsample_rates"]): #
stride_f0 = np.prod(h["upsample_rates"][i + 1 :])
self.noise_convs.append(
Conv1d(
1,
c_cur,
kernel_size=stride_f0 * 2,
stride=stride_f0,
padding=stride_f0 // 2,
)
)
else:
self.noise_convs.append(Conv1d(1, c_cur, kernel_size=1))
self.resblocks = nn.ModuleList()
for i in range(len(self.ups)):
ch = h["upsample_initial_channel"] // (2 ** (i + 1))
for j, (k, d) in enumerate(zip(h["resblock_kernel_sizes"], h["resblock_dilation_sizes"])):
self.resblocks.append(resblock(ch, k, d))
self.conv_post = weight_norm(Conv1d(ch, 1, 7, 1, padding=3))
self.ups.apply(init_weights)
self.conv_post.apply(init_weights)
self.cond = nn.Conv1d(h["gin_channels"], h["upsample_initial_channel"], 1)
[docs]
def forward(self, x, f0, g=None):
# LOG.info(1,x.shape,f0.shape,f0[:, None].shape)
f0 = self.f0_upsamp(f0[:, None]).transpose(1, 2) # bs,n,t
# LOG.info(2,f0.shape)
har_source, noi_source, uv = self.m_source(f0)
har_source = har_source.transpose(1, 2)
x = self.conv_pre(x)
x = x + self.cond(g)
# LOG.info(124,x.shape,har_source.shape)
for i in range(self.num_upsamples):
x = F.leaky_relu(x, LRELU_SLOPE)
# LOG.info(3,x.shape)
x = self.ups[i](x)
x_source = self.noise_convs[i](har_source)
# LOG.info(4,x_source.shape,har_source.shape,x.shape)
x = x + x_source
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
x = F.leaky_relu(x)
x = self.conv_post(x)
x = torch.tanh(x)
return x
[docs]
def remove_weight_norm(self):
LOG.info("Removing weight norm...")
for l in self.ups:
remove_weight_norm(l)
for l in self.resblocks:
l.remove_weight_norm()
remove_weight_norm(self.conv_pre)
remove_weight_norm(self.conv_post)