diff --git a/configs/rec/multi_language/generate_multi_language_configs.py b/configs/rec/multi_language/generate_multi_language_configs.py
index b0a87af37610f46ae7d9b87338ca8e7812ef450a..4319482b1143116de52e34562c5b9f922f0ce66c 100644
--- a/configs/rec/multi_language/generate_multi_language_configs.py
+++ b/configs/rec/multi_language/generate_multi_language_configs.py
@@ -77,6 +77,7 @@ class ArgsParser(ArgumentParser):
         global_config['Global']['save_model_dir'] = './output/rec_{}_lite'.format(type[0])
         global_config['Train']['dataset']['label_file_list'] = ["train_data/{}_train.txt".format(type[0])]
         global_config['Eval']['dataset']['label_file_list'] = ["train_data/{}_val.txt".format(type[0])]
+        global_config['Global']['character_type'] = type[0]
         assert(
                 os.path.isfile(os.path.join(project_path,global_config['Global']['character_dict_path']))
               ),"Loss default dictionary file {}_dict.txt.You can download it from \
diff --git a/configs/rec/multi_language/rec_multi_language_lite_train.yml b/configs/rec/multi_language/rec_multi_language_lite_train.yml
index 729dedd46f89af72161199d88e54417e5d206671..c42a3d1a3a5971365967effd4fdb2cc43725ef75 100644
--- a/configs/rec/multi_language/rec_multi_language_lite_train.yml
+++ b/configs/rec/multi_language/rec_multi_language_lite_train.yml
@@ -17,7 +17,7 @@ Global:
   # for data or label process
   character_dict_path: 
   # Set the language of training, if set, select the default dictionary file
-  character_type: ch
+  character_type: 
   max_text_length: 25
   infer_mode: False
   use_space_char: True
diff --git a/ppocr/utils/dict/german_dict.txt b/ppocr/utils/dict/german_dict.txt
index 6c4f7df8b199b950a48f554e189bc081e23c9578..5e121af21a1617dd970234ca98ae7072b0335332 100644
--- a/ppocr/utils/dict/german_dict.txt
+++ b/ppocr/utils/dict/german_dict.txt
@@ -137,7 +137,7 @@ z
 Ṡ
 ‘
 €
-店
-旺
-潮
-酒
+©
+ª
+«
+¬