fumiama/Retrieval-based-Voice-Conversion-WebUI
1
0
Fork 0
mirror of https://github.com/fumiama/Retrieval-based-Voice-Conversion-WebUI.git synced 2026-06-05 09:10:25 +08:00
Code Activity

chore(format): run black on dev (#11)

Browse Source 
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
This commit is contained in:
github-actions[bot]
2024-06-08 00:46:57 +09:00
committed by GitHub
parent b91dcf2261
commit 26dc9ba98d
7 changed files with 11 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
infer/lib/infer_pack/models.py
View File

@@ -355,9 +355,7 @@ class GeneratorNSF(torch.nn.Module):
self.num_upsamples = len(upsample_rates) self.num_upsamples = len(upsample_rates)
self.f0_upsamp = torch.nn.Upsample(scale_factor=math.prod(upsample_rates)) self.f0_upsamp = torch.nn.Upsample(scale_factor=math.prod(upsample_rates))
self.m_source = SourceModuleHnNSF( self.m_source = SourceModuleHnNSF(sampling_rate=sr, harmonic_num=0)
sampling_rate=sr, harmonic_num=0
)
self.noise_convs = nn.ModuleList() self.noise_convs = nn.ModuleList()
self.conv_pre = Conv1d( self.conv_pre = Conv1d(
initial_channel, upsample_initial_channel, 7, 1, padding=3 initial_channel, upsample_initial_channel, 7, 1, padding=3

2
rvc/attentions.py
View File

@@ -263,7 +263,7 @@ class FFN(nn.Module):
x = self.conv_2(self._padding(x, x_mask)) x = self.conv_2(self._padding(x, x_mask))
return x * x_mask return x * x_mask
def _padding(self, x: torch.Tensor, x_mask: torch.Tensor) -> torch.Tensor: def _padding(self, x: torch.Tensor, x_mask: torch.Tensor) -> torch.Tensor:
if self.causal: if self.causal:
return self._causal_padding(x * x_mask) return self._causal_padding(x * x_mask)

2
rvc/norms.py
View File

@@ -6,6 +6,7 @@ from torch.nn import functional as F
from .utils import activate_add_tanh_sigmoid_multiply from .utils import activate_add_tanh_sigmoid_multiply
class LayerNorm(nn.Module): class LayerNorm(nn.Module):
def __init__(self, channels: int, eps: float = 1e-5): def __init__(self, channels: int, eps: float = 1e-5):
super(LayerNorm, self).__init__() super(LayerNorm, self).__init__()
@@ -20,6 +21,7 @@ class LayerNorm(nn.Module):
x = F.layer_norm(x, (self.channels,), self.gamma, self.beta, self.eps) x = F.layer_norm(x, (self.channels,), self.gamma, self.beta, self.eps)
return x.transpose(1, -1) return x.transpose(1, -1)
class WN(torch.nn.Module): class WN(torch.nn.Module):
def __init__( def __init__(
self, self,

4
rvc/onnx/f0predictors/dio.py
View File

@@ -10,9 +10,7 @@ class DioF0Predictor(F0Predictor):
def __init__(self, hop_length=512, f0_min=50, f0_max=1100, sampling_rate=44100): def __init__(self, hop_length=512, f0_min=50, f0_max=1100, sampling_rate=44100):
super().__init__(hop_length, f0_min, f0_max, sampling_rate) super().__init__(hop_length, f0_min, f0_max, sampling_rate)
def compute_f0( def compute_f0(self, wav: np.ndarray[Any, np.dtype], p_len: Optional[int] = None):
self, wav: np.ndarray[Any, np.dtype], p_len: Optional[int] = None
):
if p_len is None: if p_len is None:
p_len = wav.shape[0] // self.hop_length p_len = wav.shape[0] // self.hop_length
f0, t = pyworld.dio( f0, t = pyworld.dio(

4
rvc/onnx/f0predictors/harvest.py
View File

@@ -10,9 +10,7 @@ class HarvestF0Predictor(F0Predictor):
def __init__(self, hop_length=512, f0_min=50, f0_max=1100, sampling_rate=44100): def __init__(self, hop_length=512, f0_min=50, f0_max=1100, sampling_rate=44100):
super().__init__(hop_length, f0_min, f0_max, sampling_rate) super().__init__(hop_length, f0_min, f0_max, sampling_rate)
def compute_f0( def compute_f0(self, wav: np.ndarray[Any, np.dtype], p_len: Optional[int] = None):
self, wav: np.ndarray[Any, np.dtype], p_len: Optional[int] = None
):
if p_len is None: if p_len is None:
p_len = wav.shape[0] // self.hop_length p_len = wav.shape[0] // self.hop_length
f0, t = pyworld.harvest( f0, t = pyworld.harvest(

4
rvc/onnx/f0predictors/pm.py
View File

@@ -10,9 +10,7 @@ class PMF0Predictor(F0Predictor):
def __init__(self, hop_length=512, f0_min=50, f0_max=1100, sampling_rate=44100): def __init__(self, hop_length=512, f0_min=50, f0_max=1100, sampling_rate=44100):
super().__init__(hop_length, f0_min, f0_max, sampling_rate) super().__init__(hop_length, f0_min, f0_max, sampling_rate)
def compute_f0( def compute_f0(self, wav: np.ndarray[Any, np.dtype], p_len: Optional[int] = None):
self, wav: np.ndarray[Any, np.dtype], p_len: Optional[int] = None
):
x = wav x = wav
if p_len is None: if p_len is None:
p_len = x.shape[0] // self.hop_length p_len = x.shape[0] // self.hop_length

4
rvc/residuals.py
View File

@@ -14,6 +14,7 @@ from .utils import (
LRELU_SLOPE = 0.1 LRELU_SLOPE = 0.1
class ResBlock1(torch.nn.Module): class ResBlock1(torch.nn.Module):
def __init__( def __init__(
self, self,
@@ -111,6 +112,7 @@ class ResBlock2(torch.nn.Module):
Actually this module is not used currently Actually this module is not used currently
because all configs specified "resblock": "1" because all configs specified "resblock": "1"
""" """
def __init__( def __init__(
self, self,
channels: int, channels: int,
@@ -255,6 +257,7 @@ class ResidualCouplingLayer(nn.Module):
torch.nn.utils.remove_weight_norm(self.enc) torch.nn.utils.remove_weight_norm(self.enc)
return self return self
class ResidualCouplingBlock(nn.Module): class ResidualCouplingBlock(nn.Module):
class Flip(nn.Module): class Flip(nn.Module):
""" """
@@ -262,6 +265,7 @@ class ResidualCouplingBlock(nn.Module):
can't take variable number of arguments or can't take variable number of arguments or
use keyword-only arguments with defaults use keyword-only arguments with defaults
""" """
def forward( def forward(
self, self,
x: torch.Tensor, x: torch.Tensor,