diff --git a/tools/infer/predict_rec.py b/tools/infer/predict_rec.py
index c8c0797b499055ec681d0362a054a30d7322b65a..a3f68f04c984c6fe50f6340d203f512cde858502 100755
--- a/tools/infer/predict_rec.py
+++ b/tools/infer/predict_rec.py
@@ -59,7 +59,7 @@ class TextRecognizer(object):
 
     def __call__(self, img_list):
         img_num = len(img_list)
-        batch_num = 30
+        batch_num = 12
         rec_res = []
         predict_time = 0
         for beg_img_no in range(0, img_num, batch_num):
diff --git a/tools/infer/predict_system.py b/tools/infer/predict_system.py
index 8b0abd706cd60a23753a4b53915a0fe5fdaa66e4..3953fa0df76c3929a52af39901f7eeb0569c68b2 100755
--- a/tools/infer/predict_system.py
+++ b/tools/infer/predict_system.py
@@ -89,7 +89,7 @@ def sorted_boxes(dt_boxes):
         sorted boxes(array) with shape [4, 2]
     """
     num_boxes = dt_boxes.shape[0]
-    sorted_boxes = sorted(dt_boxes, key=lambda x: x[0][1])
+    sorted_boxes = sorted(dt_boxes, key=lambda x: (x[0][1], x[0][0]))
     _boxes = list(sorted_boxes)
 
     for i in range(num_boxes - 1):