diff --git a/demos/streaming_asr_server/conf/ws_ds2_application.yaml b/demos/streaming_asr_server/conf/ws_ds2_application.yaml
index f67d3157..a4e6e9a1 100644
--- a/demos/streaming_asr_server/conf/ws_ds2_application.yaml
+++ b/demos/streaming_asr_server/conf/ws_ds2_application.yaml
@@ -22,8 +22,8 @@ engine_list: ['asr_online-inference']
 ################### speech task: asr; engine_type: online-inference #######################
 asr_online-inference:
     model_type: 'deepspeech2online_aishell'
-    am_model:   # the pdmodel file of am static model [optional]
-    am_params:  # the pdiparams file of am static model [optional]
+    am_model:    # the pdmodel file of am static model [optional]
+    am_params:   # the pdiparams file of am static model [optional]
     lang: 'zh'
     sample_rate: 16000
     cfg_path: 
@@ -54,7 +54,7 @@ asr_online-inference:
 ################### speech task: asr; engine_type: online-onnx #######################
 asr_online-onnx:
     model_type: 'deepspeech2online_aishell'
-    am_model:  # the pdmodel file of am static model [optional]
+    am_model:  # the pdmodel file of onnx am static model [optional]
     am_params:  # the pdiparams file of am static model [optional]
     lang: 'zh'
     sample_rate: 16000
diff --git a/paddlespeech/resource/resource.py b/paddlespeech/resource/resource.py
index 2e637f0f..15112ba7 100644
--- a/paddlespeech/resource/resource.py
+++ b/paddlespeech/resource/resource.py
@@ -168,7 +168,6 @@ class CommonTaskResource:
             exec('from .pretrained_models import {}'.format(import_models))
             models = OrderedDict(locals()[import_models])
         except Exception as e:
-            print(e)
             models = OrderedDict({})  # no models.
         finally:
             return models
diff --git a/paddlespeech/server/conf/ws_ds2_application.yaml b/paddlespeech/server/conf/ws_ds2_application.yaml
index f0a98e72..430e6fd1 100644
--- a/paddlespeech/server/conf/ws_ds2_application.yaml
+++ b/paddlespeech/server/conf/ws_ds2_application.yaml
@@ -11,7 +11,7 @@ port: 8090
 # protocol = ['websocket'] (only one can be selected).
 # websocket only support online engine type.
 protocol: 'websocket'
-engine_list: ['asr_online-onnx']
+engine_list: ['asr_online-inference']
 
 
 #################################################################################
@@ -22,8 +22,8 @@ engine_list: ['asr_online-onnx']
 ################### speech task: asr; engine_type: online-inference #######################
 asr_online-inference:
     model_type: 'deepspeech2online_aishell'
-    am_model:   # the pdmodel file of am static model [optional]
-    am_params:  # the pdiparams file of am static model [optional]
+    am_model:    # the pdmodel file of am static model [optional]
+    am_params:   # the pdiparams file of am static model [optional]
     lang: 'zh'
     sample_rate: 16000
     cfg_path: 
@@ -54,7 +54,7 @@ asr_online-inference:
 ################### speech task: asr; engine_type: online-onnx #######################
 asr_online-onnx:
     model_type: 'deepspeech2online_aishell'
-    am_model:  # the pdmodel file of am static model [optional]
+    am_model:  # the pdmodel file of onnx am static model [optional]
     am_params:  # the pdiparams file of am static model [optional]
     lang: 'zh'
     sample_rate: 16000
@@ -81,4 +81,4 @@ asr_online-onnx:
         window_n: 7     # frame
         shift_n: 4      # frame
         window_ms: 20   # ms
-        shift_ms: 10    # ms
+        shift_ms: 10    # ms
\ No newline at end of file
diff --git a/paddlespeech/server/engine/asr/online/onnx/asr_engine.py b/paddlespeech/server/engine/asr/online/onnx/asr_engine.py
index 97addc7a..aab29f78 100644
--- a/paddlespeech/server/engine/asr/online/onnx/asr_engine.py
+++ b/paddlespeech/server/engine/asr/online/onnx/asr_engine.py
@@ -331,6 +331,13 @@ class PaddleASRConnectionHanddler:
         else:
             return ''
 
+    def get_word_time_stamp(self):
+        return []
+
+    @paddle.no_grad()
+    def rescoring(self):
+        ...
+
 
 class ASRServerExecutor(ASRExecutor):
     def __init__(self):
@@ -409,17 +416,18 @@ class ASRServerExecutor(ASRExecutor):
                 os.path.dirname(os.path.abspath(self.cfg_path)))
 
         self.am_model = os.path.join(self.res_path, self.task_resource.res_dict[
-            'model']) if am_model is None else os.path.abspath(am_model)
-        self.am_params = os.path.join(
-            self.res_path, self.task_resource.res_dict[
-                'params']) if am_params is None else os.path.abspath(am_params)
+            'onnx_model']) if am_model is None else os.path.abspath(am_model)
+
+        # self.am_params = os.path.join(
+        #     self.res_path, self.task_resource.res_dict[
+        #         'params']) if am_params is None else os.path.abspath(am_params)
 
         logger.info("Load the pretrained model:")
         logger.info(f"  tag = {tag}")
         logger.info(f"  res_path: {self.res_path}")
         logger.info(f"  cfg path: {self.cfg_path}")
         logger.info(f"  am_model path: {self.am_model}")
-        logger.info(f"  am_params path: {self.am_params}")
+        # logger.info(f"  am_params path: {self.am_params}")
 
         #Init body.
         self.config = CfgNode(new_allowed=True)
diff --git a/paddlespeech/server/engine/asr/online/paddleinference/asr_engine.py b/paddlespeech/server/engine/asr/online/paddleinference/asr_engine.py
index fb24cab9..b3b31a5a 100644
--- a/paddlespeech/server/engine/asr/online/paddleinference/asr_engine.py
+++ b/paddlespeech/server/engine/asr/online/paddleinference/asr_engine.py
@@ -345,6 +345,12 @@ class PaddleASRConnectionHanddler:
         else:
             return ''
 
+    def get_word_time_stamp(self):
+        return []
+
+    @paddle.no_grad()
+    def rescoring(self):
+        ...
 
 class ASRServerExecutor(ASRExecutor):
     def __init__(self):