diff --git a/tools/infer/predict_rec.py b/tools/infer/predict_rec.py
index f754049491d8cf25a68861f0aad1c4d9eebc7e6e..e7a0111ced05b68072423ff909d93b121201fcf1 100755
--- a/tools/infer/predict_rec.py
+++ b/tools/infer/predict_rec.py
@@ -73,7 +73,7 @@ class TextRecognizer(object):
         width_list = []
         for img in img_list:
             width_list.append(img.shape[1] / float(img.shape[0]))
-        # Sorting can be accelerated
+        # Sorting can speed up the recognition process
         indices = np.argsort(np.array(width_list))
 
         # rec_res = []