diff --git a/nemo/collections/asr/data/audio_to_text.py b/nemo/collections/asr/data/audio_to_text.py
index 4d0ad776841b..1e367fd0bf51 100644
--- a/nemo/collections/asr/data/audio_to_text.py
+++ b/nemo/collections/asr/data/audio_to_text.py
@@ -1195,7 +1195,7 @@ def __init__(self, tokenizer):
                 self._tokenizer = tokenizer
 
             def __call__(self, *args):
-                if isinstance(args[0], Iterable) and self.is_aggregate:
+                if isinstance(args[0], List) and self.is_aggregate:
                     t = []
                     for span in args[0]:
                         t.extend(self._tokenizer.text_to_ids(span['str'], span['lang']))