pass x_mask to layer norm

TTS/vocoder/layers/pqmf.py

@@ -22,7 +22,7 @@ class PQMF(torch.nn.Module):
         for k in range(N):
             constant_factor = (2 * k + 1) * (np.pi /
                                              (2 * N)) * (np.arange(taps + 1) -
-                                                         ((taps - 1) / 2))
+                                                         ((taps - 1) / 2))  # TODO: (taps - 1) -> taps
             phase = (-1)**k * np.pi / 4
             H[k] = 2 * QMF * np.cos(constant_factor + phase)
 

mozilla_voice_tts/tts/layers/glow_tts/encoder.py

@@ -121,7 +121,7 @@ class Encoder(nn.Module):
                                  1).to(x.dtype)
         # pass encoder
         for layer in self.encoder:
-            x = layer(x)
+            x = layer(x, x_mask)
         # set duration predictor input
         if g is not None:
             g_exp = g.expand(-1, -1, x.size(-1))