diff --git a/data_utils.py b/data_utils.py
index 7f4bfbe..af7d601 100644
--- a/data_utils.py
+++ b/data_utils.py
@@ -208,7 +208,7 @@ class TextAudioSpeakerCollate:
             torch.LongTensor([x[1].size(1) for x in batch]), dim=0, descending=True
         )
 
-        max_text_len = max([len(x[0]) for x in batch])
+        max_text_len = max([batch[ids_sorted_decreasing[i]][7].size(1) for i in range(len(ids_sorted_decreasing))] + [len(x[0]) for x in batch])
         max_spec_len = max([x[1].size(1) for x in batch])
         max_wav_len = max([x[2].size(1) for x in batch])