tacotron2-gst-en / loss_function.py
mireiafarrus's picture
tacotron2 and hifigan upload
af7ac2b
raw
history blame contribute delete
No virus
1.22 kB
from torch import nn
class Tacotron2Loss(nn.Module):
def __init__(self):
super(Tacotron2Loss, self).__init__()
def forward(self, model_output, targets):
mel_target, gate_target = targets[0], targets[1]
mel_target.requires_grad = False
gate_target.requires_grad = False
# Ensures dimension 1 will be size 1, the rest can be adapted. It is a column of length 189 with all zeroes
# till the end of the current sequence, which is filled with 1's
gate_target = gate_target.view(-1, 1)
mel_out, mel_out_postnet, gate_out, _, _ = model_output
gate_out = gate_out.view(-1, 1)
# Mean Square Error (L2) loss function for decoder generation + post net generation
mel_loss = nn.MSELoss()(mel_out, mel_target) + \
nn.MSELoss()(mel_out_postnet, mel_target)
# Binary Cross Entropy with a Sigmoid layer combined. It is more efficient than using a plain Sigmoid
# followed by a BCELoss as, by combining the operations into one layer, we take advantage of the log-sum-exp
# trick for numerical stability
gate_loss = nn.BCEWithLogitsLoss()(gate_out, gate_target)
return mel_loss + gate_loss