You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
PaddleSpeech/paddlespeech/t2s/models/hifigan/hifigan.py

780 lines
27 KiB

# -*- coding: utf-8 -*-
"""HiFi-GAN Modules.
This code is based on https://github.com/jik876/hifi-gan.
"""
import copy
from typing import Any
from typing import Dict
from typing import List
import paddle
import paddle.nn.functional as F
from paddle import nn
from paddlespeech.t2s.modules.activation import get_activation
from paddlespeech.t2s.modules.nets_utils import initialize
from paddlespeech.t2s.modules.residual_block import HiFiGANResidualBlock as ResidualBlock
class HiFiGANGenerator(nn.Layer):
"""HiFiGAN generator module."""
def __init__(
self,
in_channels: int=80,
out_channels: int=1,
channels: int=512,
kernel_size: int=7,
upsample_scales: List[int]=(8, 8, 2, 2),
upsample_kernel_sizes: List[int]=(16, 16, 4, 4),
resblock_kernel_sizes: List[int]=(3, 7, 11),
resblock_dilations: List[List[int]]=[(1, 3, 5), (1, 3, 5),
(1, 3, 5)],
use_additional_convs: bool=True,
bias: bool=True,
nonlinear_activation: str="leakyrelu",
nonlinear_activation_params: Dict[str, Any]={"negative_slope": 0.1},
use_weight_norm: bool=True,
init_type: str="xavier_uniform", ):
"""Initialize HiFiGANGenerator module.
Parameters
----------
in_channels : int
Number of input channels.
out_channels : int
Number of output channels.
channels : int
Number of hidden representation channels.
kernel_size : int
Kernel size of initial and final conv layer.
upsample_scales : list
List of upsampling scales.
upsample_kernel_sizes : list
List of kernel sizes for upsampling layers.
resblock_kernel_sizes : list
List of kernel sizes for residual blocks.
resblock_dilations : list
List of dilation list for residual blocks.
use_additional_convs : bool
Whether to use additional conv layers in residual blocks.
bias : bool
Whether to add bias parameter in convolution layers.
nonlinear_activation : str
Activation function module name.
nonlinear_activation_params : dict
Hyperparameters for activation function.
use_weight_norm : bool
Whether to use weight norm.
If set to true, it will be applied to all of the conv layers.
"""
super().__init__()
# initialize parameters
initialize(self, init_type)
# check hyperparameters are valid
assert kernel_size % 2 == 1, "Kernel size must be odd number."
assert len(upsample_scales) == len(upsample_kernel_sizes)
assert len(resblock_dilations) == len(resblock_kernel_sizes)
# define modules
self.num_upsamples = len(upsample_kernel_sizes)
self.num_blocks = len(resblock_kernel_sizes)
self.input_conv = nn.Conv1D(
in_channels,
channels,
kernel_size,
1,
padding=(kernel_size - 1) // 2, )
self.upsamples = nn.LayerList()
self.blocks = nn.LayerList()
for i in range(len(upsample_kernel_sizes)):
assert upsample_kernel_sizes[i] == 2 * upsample_scales[i]
self.upsamples.append(
nn.Sequential(
get_activation(nonlinear_activation, **
nonlinear_activation_params),
nn.Conv1DTranspose(
channels // (2**i),
channels // (2**(i + 1)),
upsample_kernel_sizes[i],
upsample_scales[i],
padding=upsample_scales[i] // 2 + upsample_scales[i] %
2,
output_padding=upsample_scales[i] % 2, ), ))
for j in range(len(resblock_kernel_sizes)):
self.blocks.append(
ResidualBlock(
kernel_size=resblock_kernel_sizes[j],
channels=channels // (2**(i + 1)),
dilations=resblock_dilations[j],
bias=bias,
use_additional_convs=use_additional_convs,
nonlinear_activation=nonlinear_activation,
nonlinear_activation_params=nonlinear_activation_params,
))
self.output_conv = nn.Sequential(
nn.LeakyReLU(),
nn.Conv1D(
channels // (2**(i + 1)),
out_channels,
kernel_size,
1,
padding=(kernel_size - 1) // 2, ),
nn.Tanh(), )
nn.initializer.set_global_initializer(None)
# apply weight norm
if use_weight_norm:
self.apply_weight_norm()
# reset parameters
self.reset_parameters()
def forward(self, c):
"""Calculate forward propagation.
Parameters
----------
c : Tensor
Input tensor (B, in_channels, T).
Returns
----------
Tensor
Output tensor (B, out_channels, T).
"""
c = self.input_conv(c)
for i in range(self.num_upsamples):
c = self.upsamples[i](c)
# initialize
cs = 0.0
for j in range(self.num_blocks):
cs += self.blocks[i * self.num_blocks + j](c)
c = cs / self.num_blocks
c = self.output_conv(c)
return c
def reset_parameters(self):
"""Reset parameters.
This initialization follows official implementation manner.
https://github.com/jik876/hifi-gan/blob/master/models.py
"""
# 定义参数为float的正态分布。
dist = paddle.distribution.Normal(loc=0.0, scale=0.01)
def _reset_parameters(m):
if isinstance(m, nn.Conv1D) or isinstance(m, nn.Conv1DTranspose):
w = dist.sample(m.weight.shape)
m.weight.set_value(w)
self.apply(_reset_parameters)
def apply_weight_norm(self):
"""Recursively apply weight normalization to all the Convolution layers
in the sublayers.
"""
def _apply_weight_norm(layer):
if isinstance(layer, (nn.Conv1D, nn.Conv2D, nn.Conv1DTranspose)):
nn.utils.weight_norm(layer)
self.apply(_apply_weight_norm)
def remove_weight_norm(self):
"""Recursively remove weight normalization from all the Convolution
layers in the sublayers.
"""
def _remove_weight_norm(layer):
try:
nn.utils.remove_weight_norm(layer)
except ValueError:
pass
self.apply(_remove_weight_norm)
def inference(self, c):
"""Perform inference.
Parameters
----------
c : Tensor
Input tensor (T, in_channels).
normalize_before (bool): Whether to perform normalization.
Returns
----------
Tensor
Output tensor (T ** prod(upsample_scales), out_channels).
"""
c = self.forward(c.transpose([1, 0]).unsqueeze(0))
return c.squeeze(0).transpose([1, 0])
class HiFiGANPeriodDiscriminator(nn.Layer):
"""HiFiGAN period discriminator module."""
def __init__(
self,
in_channels: int=1,
out_channels: int=1,
period: int=3,
kernel_sizes: List[int]=[5, 3],
channels: int=32,
downsample_scales: List[int]=[3, 3, 3, 3, 1],
max_downsample_channels: int=1024,
bias: bool=True,
nonlinear_activation: str="leakyrelu",
nonlinear_activation_params: Dict[str, Any]={"negative_slope": 0.1},
use_weight_norm: bool=True,
use_spectral_norm: bool=False,
init_type: str="xavier_uniform", ):
"""Initialize HiFiGANPeriodDiscriminator module.
Parameters
----------
in_channels : int
Number of input channels.
out_channels : int
Number of output channels.
period : int
Period.
kernel_sizes : list
Kernel sizes of initial conv layers and the final conv layer.
channels : int
Number of initial channels.
downsample_scales : list
List of downsampling scales.
max_downsample_channels : int
Number of maximum downsampling channels.
use_additional_convs : bool
Whether to use additional conv layers in residual blocks.
bias : bool
Whether to add bias parameter in convolution layers.
nonlinear_activation : str
Activation function module name.
nonlinear_activation_params : dict
Hyperparameters for activation function.
use_weight_norm : bool
Whether to use weight norm.
If set to true, it will be applied to all of the conv layers.
use_spectral_norm : bool
Whether to use spectral norm.
If set to true, it will be applied to all of the conv layers.
"""
super().__init__()
# initialize parameters
initialize(self, init_type)
assert len(kernel_sizes) == 2
assert kernel_sizes[0] % 2 == 1, "Kernel size must be odd number."
assert kernel_sizes[1] % 2 == 1, "Kernel size must be odd number."
self.period = period
self.convs = nn.LayerList()
in_chs = in_channels
out_chs = channels
for downsample_scale in downsample_scales:
self.convs.append(
nn.Sequential(
nn.Conv2D(
in_chs,
out_chs,
(kernel_sizes[0], 1),
(downsample_scale, 1),
padding=((kernel_sizes[0] - 1) // 2, 0), ),
get_activation(nonlinear_activation, **
nonlinear_activation_params), ))
in_chs = out_chs
# NOTE: Use downsample_scale + 1?
out_chs = min(out_chs * 4, max_downsample_channels)
self.output_conv = nn.Conv2D(
out_chs,
out_channels,
(kernel_sizes[1] - 1, 1),
1,
padding=((kernel_sizes[1] - 1) // 2, 0), )
if use_weight_norm and use_spectral_norm:
raise ValueError("Either use use_weight_norm or use_spectral_norm.")
# apply weight norm
if use_weight_norm:
self.apply_weight_norm()
# apply spectral norm
if use_spectral_norm:
self.apply_spectral_norm()
def forward(self, x):
"""Calculate forward propagation.
Parameters
----------
c : Tensor
Input tensor (B, in_channels, T).
Returns
----------
list
List of each layer's tensors.
"""
# transform 1d to 2d -> (B, C, T/P, P)
b, c, t = paddle.shape(x)
if t % self.period != 0:
n_pad = self.period - (t % self.period)
x = F.pad(x, (0, n_pad), "reflect", data_format="NCL")
t += n_pad
x = x.reshape([b, c, t // self.period, self.period])
# forward conv
outs = []
for layer in self.convs:
x = layer(x)
outs += [x]
x = self.output_conv(x)
x = paddle.flatten(x, 1, -1)
outs += [x]
return outs
def apply_weight_norm(self):
"""Recursively apply weight normalization to all the Convolution layers
in the sublayers.
"""
def _apply_weight_norm(layer):
if isinstance(layer, (nn.Conv1D, nn.Conv2D, nn.Conv1DTranspose)):
nn.utils.weight_norm(layer)
self.apply(_apply_weight_norm)
def apply_spectral_norm(self):
"""Apply spectral normalization module from all of the layers."""
def _apply_spectral_norm(m):
if isinstance(m, nn.Conv2D):
nn.utils.spectral_norm(m)
self.apply(_apply_spectral_norm)
class HiFiGANMultiPeriodDiscriminator(nn.Layer):
"""HiFiGAN multi-period discriminator module."""
def __init__(
self,
periods: List[int]=[2, 3, 5, 7, 11],
discriminator_params: Dict[str, Any]={
"in_channels": 1,
"out_channels": 1,
"kernel_sizes": [5, 3],
"channels": 32,
"downsample_scales": [3, 3, 3, 3, 1],
"max_downsample_channels": 1024,
"bias": True,
"nonlinear_activation": "leakyrelu",
"nonlinear_activation_params": {
"negative_slope": 0.1
},
"use_weight_norm": True,
"use_spectral_norm": False,
},
init_type: str="xavier_uniform", ):
"""Initialize HiFiGANMultiPeriodDiscriminator module.
Parameters
----------
periods : list
List of periods.
discriminator_params : dict
Parameters for hifi-gan period discriminator module.
The period parameter will be overwritten.
"""
super().__init__()
# initialize parameters
initialize(self, init_type)
self.discriminators = nn.LayerList()
for period in periods:
params = copy.deepcopy(discriminator_params)
params["period"] = period
self.discriminators.append(HiFiGANPeriodDiscriminator(**params))
def forward(self, x):
"""Calculate forward propagation.
Parameters
----------
x : Tensor
Input noise signal (B, 1, T).
Returns
----------
List
List of list of each discriminator outputs, which consists of each layer output tensors.
"""
outs = []
for f in self.discriminators:
outs += [f(x)]
return outs
class HiFiGANScaleDiscriminator(nn.Layer):
"""HiFi-GAN scale discriminator module."""
def __init__(
self,
in_channels: int=1,
out_channels: int=1,
kernel_sizes: List[int]=[15, 41, 5, 3],
channels: int=128,
max_downsample_channels: int=1024,
max_groups: int=16,
bias: bool=True,
downsample_scales: List[int]=[2, 2, 4, 4, 1],
nonlinear_activation: str="leakyrelu",
nonlinear_activation_params: Dict[str, Any]={"negative_slope": 0.1},
use_weight_norm: bool=True,
use_spectral_norm: bool=False,
init_type: str="xavier_uniform", ):
"""Initilize HiFiGAN scale discriminator module.
Parameters
----------
in_channels : int
Number of input channels.
out_channels : int
Number of output channels.
kernel_sizes : list
List of four kernel sizes. The first will be used for the first conv layer,
and the second is for downsampling part, and the remaining two are for output layers.
channels : int
Initial number of channels for conv layer.
max_downsample_channels : int
Maximum number of channels for downsampling layers.
bias : bool
Whether to add bias parameter in convolution layers.
downsample_scales : list
List of downsampling scales.
nonlinear_activation : str
Activation function module name.
nonlinear_activation_params : dict
Hyperparameters for activation function.
use_weight_norm : bool
Whether to use weight norm.
If set to true, it will be applied to all of the conv layers.
use_spectral_norm : bool
Whether to use spectral norm.
If set to true, it will be applied to all of the conv layers.
"""
super().__init__()
# initialize parameters
initialize(self, init_type)
self.layers = nn.LayerList()
# check kernel size is valid
assert len(kernel_sizes) == 4
for ks in kernel_sizes:
assert ks % 2 == 1
# add first layer
self.layers.append(
nn.Sequential(
nn.Conv1D(
in_channels,
channels,
# NOTE: Use always the same kernel size
kernel_sizes[0],
bias_attr=bias,
padding=(kernel_sizes[0] - 1) // 2, ),
get_activation(nonlinear_activation, **
nonlinear_activation_params), ))
# add downsample layers
in_chs = channels
out_chs = channels
# NOTE(kan-bayashi): Remove hard coding?
groups = 4
for downsample_scale in downsample_scales:
self.layers.append(
nn.Sequential(
nn.Conv1D(
in_chs,
out_chs,
kernel_size=kernel_sizes[1],
stride=downsample_scale,
padding=(kernel_sizes[1] - 1) // 2,
groups=groups,
bias_attr=bias, ),
get_activation(nonlinear_activation, **
nonlinear_activation_params), ))
in_chs = out_chs
# NOTE: Remove hard coding?
out_chs = min(in_chs * 2, max_downsample_channels)
# NOTE: Remove hard coding?
groups = min(groups * 4, max_groups)
# add final layers
out_chs = min(in_chs * 2, max_downsample_channels)
self.layers.append(
nn.Sequential(
nn.Conv1D(
in_chs,
out_chs,
kernel_size=kernel_sizes[2],
stride=1,
padding=(kernel_sizes[2] - 1) // 2,
bias_attr=bias, ),
get_activation(nonlinear_activation, **
nonlinear_activation_params), ))
self.layers.append(
nn.Conv1D(
out_chs,
out_channels,
kernel_size=kernel_sizes[3],
stride=1,
padding=(kernel_sizes[3] - 1) // 2,
bias_attr=bias, ), )
if use_weight_norm and use_spectral_norm:
raise ValueError("Either use use_weight_norm or use_spectral_norm.")
# apply weight norm
if use_weight_norm:
self.apply_weight_norm()
# apply spectral norm
if use_spectral_norm:
self.apply_spectral_norm()
def forward(self, x):
"""Calculate forward propagation.
Parameters
----------
x : Tensor
Input noise signal (B, 1, T).
Returns
----------
List
List of output tensors of each layer.
"""
outs = []
for f in self.layers:
x = f(x)
outs += [x]
return outs
def apply_weight_norm(self):
"""Recursively apply weight normalization to all the Convolution layers
in the sublayers.
"""
def _apply_weight_norm(layer):
if isinstance(layer, (nn.Conv1D, nn.Conv2D, nn.Conv1DTranspose)):
nn.utils.weight_norm(layer)
self.apply(_apply_weight_norm)
def apply_spectral_norm(self):
"""Apply spectral normalization module from all of the layers."""
def _apply_spectral_norm(m):
if isinstance(m, nn.Conv2D):
nn.utils.spectral_norm(m)
self.apply(_apply_spectral_norm)
class HiFiGANMultiScaleDiscriminator(nn.Layer):
"""HiFi-GAN multi-scale discriminator module."""
def __init__(
self,
scales: int=3,
downsample_pooling: str="AvgPool1D",
# follow the official implementation setting
downsample_pooling_params: Dict[str, Any]={
"kernel_size": 4,
"stride": 2,
"padding": 2,
},
discriminator_params: Dict[str, Any]={
"in_channels": 1,
"out_channels": 1,
"kernel_sizes": [15, 41, 5, 3],
"channels": 128,
"max_downsample_channels": 1024,
"max_groups": 16,
"bias": True,
"downsample_scales": [2, 2, 4, 4, 1],
"nonlinear_activation": "leakyrelu",
"nonlinear_activation_params": {
"negative_slope": 0.1
},
},
follow_official_norm: bool=False,
init_type: str="xavier_uniform", ):
"""Initilize HiFiGAN multi-scale discriminator module.
Parameters
----------
scales : int
Number of multi-scales.
downsample_pooling : str
Pooling module name for downsampling of the inputs.
downsample_pooling_params : dict
Parameters for the above pooling module.
discriminator_params : dict
Parameters for hifi-gan scale discriminator module.
follow_official_norm : bool
Whether to follow the norm setting of the official
implementaion. The first discriminator uses spectral norm and the other
discriminators use weight norm.
"""
super().__init__()
# initialize parameters
initialize(self, init_type)
self.discriminators = nn.LayerList()
# add discriminators
for i in range(scales):
params = copy.deepcopy(discriminator_params)
if follow_official_norm:
if i == 0:
params["use_weight_norm"] = False
params["use_spectral_norm"] = True
else:
params["use_weight_norm"] = True
params["use_spectral_norm"] = False
self.discriminators.append(HiFiGANScaleDiscriminator(**params))
self.pooling = getattr(nn, downsample_pooling)(
**downsample_pooling_params)
def forward(self, x):
"""Calculate forward propagation.
Parameters
----------
x : Tensor
Input noise signal (B, 1, T).
Returns
----------
List
List of list of each discriminator outputs, which consists of each layer output tensors.
"""
outs = []
for f in self.discriminators:
outs += [f(x)]
x = self.pooling(x)
return outs
class HiFiGANMultiScaleMultiPeriodDiscriminator(nn.Layer):
"""HiFi-GAN multi-scale + multi-period discriminator module."""
def __init__(
self,
# Multi-scale discriminator related
scales: int=3,
scale_downsample_pooling: str="AvgPool1D",
scale_downsample_pooling_params: Dict[str, Any]={
"kernel_size": 4,
"stride": 2,
"padding": 2,
},
scale_discriminator_params: Dict[str, Any]={
"in_channels": 1,
"out_channels": 1,
"kernel_sizes": [15, 41, 5, 3],
"channels": 128,
"max_downsample_channels": 1024,
"max_groups": 16,
"bias": True,
"downsample_scales": [2, 2, 4, 4, 1],
"nonlinear_activation": "leakyrelu",
"nonlinear_activation_params": {
"negative_slope": 0.1
},
},
follow_official_norm: bool=True,
# Multi-period discriminator related
periods: List[int]=[2, 3, 5, 7, 11],
period_discriminator_params: Dict[str, Any]={
"in_channels": 1,
"out_channels": 1,
"kernel_sizes": [5, 3],
"channels": 32,
"downsample_scales": [3, 3, 3, 3, 1],
"max_downsample_channels": 1024,
"bias": True,
"nonlinear_activation": "leakyrelu",
"nonlinear_activation_params": {
"negative_slope": 0.1
},
"use_weight_norm": True,
"use_spectral_norm": False,
},
init_type: str="xavier_uniform", ):
"""Initilize HiFiGAN multi-scale + multi-period discriminator module.
Parameters
----------
scales : int
Number of multi-scales.
scale_downsample_pooling : str
Pooling module name for downsampling of the inputs.
scale_downsample_pooling_params : dict
Parameters for the above pooling module.
scale_discriminator_params : dict
Parameters for hifi-gan scale discriminator module.
follow_official_norm : bool): Whether to follow the norm setting of the official
implementaion. The first discriminator uses spectral norm and the other
discriminators use weight norm.
periods : list
List of periods.
period_discriminator_params : dict
Parameters for hifi-gan period discriminator module.
The period parameter will be overwritten.
"""
super().__init__()
# initialize parameters
initialize(self, init_type)
self.msd = HiFiGANMultiScaleDiscriminator(
scales=scales,
downsample_pooling=scale_downsample_pooling,
downsample_pooling_params=scale_downsample_pooling_params,
discriminator_params=scale_discriminator_params,
follow_official_norm=follow_official_norm, )
self.mpd = HiFiGANMultiPeriodDiscriminator(
periods=periods,
discriminator_params=period_discriminator_params, )
def forward(self, x):
"""Calculate forward propagation.
Parameters
----------
x : Tensor
Input noise signal (B, 1, T).
Returns
----------
List:
List of list of each discriminator outputs,
which consists of each layer output tensors.
Multi scale and multi period ones are concatenated.
"""
msd_outs = self.msd(x)
mpd_outs = self.mpd(x)
return msd_outs + mpd_outs
class HiFiGANInference(nn.Layer):
def __init__(self, normalizer, hifigan_generator):
super().__init__()
self.normalizer = normalizer
self.hifigan_generator = hifigan_generator
def forward(self, logmel):
normalized_mel = self.normalizer(logmel)
wav = self.hifigan_generator.inference(normalized_mel)
return wav