diff --git a/config.json b/config.json
index 837aa167..c41b1aa3 100644
--- a/config.json
+++ b/config.json
@@ -39,6 +39,7 @@
     "warmup_steps": 4000,   // Noam decay steps to increase the learning rate from 0 to "lr"
     "windowing": false,      // Enables attention windowing. Used only in eval mode.
     "memory_size": 5,       // TO BE IMPLEMENTED -- memory queue size used to queue network predictions to feed autoregressive connection. Useful if r < 5. 
+    "attention_norm": "softmax",   // softmax or sigmoid. Suggested to use softmax for Tacotron2 and sigmoid for Tacotron.
 
     "batch_size": 16,       // Batch size for training. Lower values than 32 might cause hard to learn attention.
     "eval_batch_size":16,   
diff --git a/models/tacotron.py b/models/tacotron.py
index 7bda5ea2..adf74ab3 100644
--- a/models/tacotron.py
+++ b/models/tacotron.py
@@ -14,7 +14,7 @@ class Tacotron(nn.Module):
                  r=5,
                  padding_idx=None,
                  memory_size=5,
-                 attn_windowing=False,
+                 attn_win=False,
                  attn_norm="sigmoid"):
         super(Tacotron, self).__init__()
         self.r = r
@@ -23,7 +23,7 @@ class Tacotron(nn.Module):
         self.embedding = nn.Embedding(num_chars, 256, padding_idx=padding_idx)
         self.embedding.weight.data.normal_(0, 0.3)
         self.encoder = Encoder(256)
-        self.decoder = Decoder(256, mel_dim, r, memory_size, attn_windowing, attn_norm)
+        self.decoder = Decoder(256, mel_dim, r, memory_size, attn_win, attn_norm)
         self.postnet = PostCBHG(mel_dim)
         self.last_linear = nn.Sequential(
             nn.Linear(self.postnet.cbhg.gru_features * 2, linear_dim),
diff --git a/train.py b/train.py
index e4e4a674..83836bba 100644
--- a/train.py
+++ b/train.py
@@ -375,7 +375,7 @@ def main(args):
         init_distributed(args.rank, num_gpus, args.group_id,
                          c.distributed["backend"], c.distributed["url"])
     num_chars = len(phonemes) if c.use_phonemes else len(symbols)
-    model = MyModel(num_chars=num_chars, r=c.r, attention_norm=c.attention_norm)
+    model = MyModel(num_chars=num_chars, r=c.r, attn_norm=c.attention_norm)
 
     print(" | > Num output units : {}".format(ap.num_freq), flush=True)