OPENGIT/Retrieval-based-Voice-Conversion-WebUI
1
0
Fork 0
mirror of https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI.git synced 2025-05-08 04:51:44 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix the receptive field of flow

Browse Source
This commit is contained in:
yxlllc 2024-03-06 16:20:50 +08:00
parent 7728caad54
commit 8e78f655e6
1 changed files with 9 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
infer/lib/infer_pack/models.py
View File

@ -794,15 +794,15 @@ class SynthesizerTrnMs256NSFsid(nn.Module):
g = self.emb_g(sid).unsqueeze(-1) g = self.emb_g(sid).unsqueeze(-1)
m_p, logs_p, x_mask = self.enc_p(phone, pitch, phone_lengths) m_p, logs_p, x_mask = self.enc_p(phone, pitch, phone_lengths)
z_p = (m_p + torch.exp(logs_p) * torch.randn_like(m_p) * 0.66666) * x_mask z_p = (m_p + torch.exp(logs_p) * torch.randn_like(m_p) * 0.66666) * x_mask
z = self.flow(z_p, x_mask, g=g, reverse=True)
if skip_head is not None and return_length is not None: if skip_head is not None and return_length is not None:
assert isinstance(skip_head, torch.Tensor) assert isinstance(skip_head, torch.Tensor)
assert isinstance(return_length, torch.Tensor) assert isinstance(return_length, torch.Tensor)
head = int(skip_head.item()) head = int(skip_head.item())
length = int(return_length.item()) length = int(return_length.item())
z_p = z_p[:, :, head : head + length] z = z[:, :, head : head + length]
x_mask = x_mask[:, :, head : head + length] x_mask = x_mask[:, :, head : head + length]
nsff0 = nsff0[:, head : head + length] nsff0 = nsff0[:, head : head + length]
z = self.flow(z_p, x_mask, g=g, reverse=True)
o = self.dec(z * x_mask, nsff0, g=g) o = self.dec(z * x_mask, nsff0, g=g)
return o, x_mask, (z, z_p, m_p, logs_p) return o, x_mask, (z, z_p, m_p, logs_p)
@ -956,15 +956,15 @@ class SynthesizerTrnMs768NSFsid(nn.Module):
g = self.emb_g(sid).unsqueeze(-1) g = self.emb_g(sid).unsqueeze(-1)
m_p, logs_p, x_mask = self.enc_p(phone, pitch, phone_lengths) m_p, logs_p, x_mask = self.enc_p(phone, pitch, phone_lengths)
z_p = (m_p + torch.exp(logs_p) * torch.randn_like(m_p) * 0.66666) * x_mask z_p = (m_p + torch.exp(logs_p) * torch.randn_like(m_p) * 0.66666) * x_mask
z = self.flow(z_p, x_mask, g=g, reverse=True)
if skip_head is not None and return_length is not None: if skip_head is not None and return_length is not None:
assert isinstance(skip_head, torch.Tensor) assert isinstance(skip_head, torch.Tensor)
assert isinstance(return_length, torch.Tensor) assert isinstance(return_length, torch.Tensor)
head = int(skip_head.item()) head = int(skip_head.item())
length = int(return_length.item()) length = int(return_length.item())
z_p = z_p[:, :, head : head + length] z = z[:, :, head : head + length]
x_mask = x_mask[:, :, head : head + length] x_mask = x_mask[:, :, head : head + length]
nsff0 = nsff0[:, head : head + length] nsff0 = nsff0[:, head : head + length]
z = self.flow(z_p, x_mask, g=g, reverse=True)
o = self.dec(z * x_mask, nsff0, g=g) o = self.dec(z * x_mask, nsff0, g=g)
return o, x_mask, (z, z_p, m_p, logs_p) return o, x_mask, (z, z_p, m_p, logs_p)
@ -1107,14 +1107,14 @@ class SynthesizerTrnMs256NSFsid_nono(nn.Module):
g = self.emb_g(sid).unsqueeze(-1) g = self.emb_g(sid).unsqueeze(-1)
m_p, logs_p, x_mask = self.enc_p(phone, None, phone_lengths) m_p, logs_p, x_mask = self.enc_p(phone, None, phone_lengths)
z_p = (m_p + torch.exp(logs_p) * torch.randn_like(m_p) * 0.66666) * x_mask z_p = (m_p + torch.exp(logs_p) * torch.randn_like(m_p) * 0.66666) * x_mask
z = self.flow(z_p, x_mask, g=g, reverse=True)
if skip_head is not None and return_length is not None: if skip_head is not None and return_length is not None:
assert isinstance(skip_head, torch.Tensor) assert isinstance(skip_head, torch.Tensor)
assert isinstance(return_length, torch.Tensor) assert isinstance(return_length, torch.Tensor)
head = int(skip_head.item()) head = int(skip_head.item())
length = int(return_length.item()) length = int(return_length.item())
z_p = z_p[:, :, head : head + length] z = z[:, :, head : head + length]
x_mask = x_mask[:, :, head : head + length] x_mask = x_mask[:, :, head : head + length]
z = self.flow(z_p, x_mask, g=g, reverse=True)
o = self.dec(z * x_mask, g=g) o = self.dec(z * x_mask, g=g)
return o, x_mask, (z, z_p, m_p, logs_p) return o, x_mask, (z, z_p, m_p, logs_p)
@ -1257,14 +1257,14 @@ class SynthesizerTrnMs768NSFsid_nono(nn.Module):
g = self.emb_g(sid).unsqueeze(-1) g = self.emb_g(sid).unsqueeze(-1)
m_p, logs_p, x_mask = self.enc_p(phone, None, phone_lengths) m_p, logs_p, x_mask = self.enc_p(phone, None, phone_lengths)
z_p = (m_p + torch.exp(logs_p) * torch.randn_like(m_p) * 0.66666) * x_mask z_p = (m_p + torch.exp(logs_p) * torch.randn_like(m_p) * 0.66666) * x_mask
z = self.flow(z_p, x_mask, g=g, reverse=True)
if skip_head is not None and return_length is not None: if skip_head is not None and return_length is not None:
assert isinstance(skip_head, torch.Tensor) assert isinstance(skip_head, torch.Tensor)
assert isinstance(return_length, torch.Tensor) assert isinstance(return_length, torch.Tensor)
head = int(skip_head.item()) head = int(skip_head.item())
length = int(return_length.item()) length = int(return_length.item())
z_p = z_p[:, :, head : head + length] z = z[:, :, head : head + length]
x_mask = x_mask[:, :, head : head + length] x_mask = x_mask[:, :, head : head + length]
z = self.flow(z_p, x_mask, g=g, reverse=True)
o = self.dec(z * x_mask, g=g) o = self.dec(z * x_mask, g=g)
return o, x_mask, (z, z_p, m_p, logs_p) return o, x_mask, (z, z_p, m_p, logs_p)