blackteay/GPT-SoVITS
1
0
Fork 0
mirror of https://github.com/RVC-Boss/GPT-SoVITS.git synced 2025-10-06 06:29:59 +08:00
Code Issues Packages Projects Releases Wiki Activity

remove GB code

Browse Source
This commit is contained in:
starylan 2025-02-12 01:46:30 +08:00
parent 05cffb56ca
commit 4748f1a6b0
1 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
GPT_SoVITS/module/models.py
View File

@ -243,7 +243,7 @@ class TextEncoder(nn.Module):
)
y = self.ssl_proj(y * y_mask) * y_mask
y = self.encoder_ssl(y * y_mask, y_mask)
text_mask = torch.unsqueeze(
@ -1139,8 +1139,8 @@ class CFM(torch.nn.Module):
class SynthesizerTrnV3(nn.Module):
"""
Synthesizer for Training
"""
Synthesizer for Training
"""
def __init__(self,
spec_channels,
@ -1187,8 +1187,8 @@ class SynthesizerTrnV3(nn.Module):
self.model_dim=512
self.use_sdp = use_sdp
self.enc_p = TextEncoder(inter_channels,hidden_channels,filter_channels,n_heads,n_layers,kernel_size,p_dropout)
# self.ref_enc = modules.MelStyleEncoder(spec_channels, style_vector_dim=gin_channels)###回滚。。。
self.ref_enc = modules.MelStyleEncoder(704, style_vector_dim=gin_channels)###回滚。。。
# self.ref_enc = modules.MelStyleEncoder(spec_channels, style_vector_dim=gin_channels)###<EFBFBD>ع<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
self.ref_enc = modules.MelStyleEncoder(704, style_vector_dim=gin_channels)###<EFBFBD>ع<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
# self.dec = Generator(inter_channels, resblock, resblock_kernel_sizes, resblock_dilation_sizes, upsample_rates,
# upsample_initial_channel, upsample_kernel_sizes, gin_channels=gin_channels)
# self.enc_q = PosteriorEncoder(spec_channels, inter_channels, hidden_channels, 5, 1, 16,
@ -1238,7 +1238,7 @@ class SynthesizerTrnV3(nn.Module):
x, m_p, logs_p, y_mask = self.enc_p(quantized, y_lengths, text, text_lengths, ge)
fea=self.bridge(x)
fea = F.interpolate(fea, scale_factor=1.875, mode="nearest")##BCT
fea, y_mask_ = self.wns1(fea, mel_lengths, ge)###如果1min微调没问题就不需要微操学习率了
fea, y_mask_ = self.wns1(fea, mel_lengths, ge)##If the 1-minute fine-tuning works fine, no need to manually adjust the learning rate.
B=ssl.shape[0]
prompt_len_max = mel_lengths*2/3
prompt_len = (torch.rand([B], device=fea.device) * prompt_len_max).floor().to(dtype=torch.long)