diff --git a/mindinsight/mindconverter/README.md b/mindinsight/mindconverter/README.md
index f2336682..3eed7390 100644
--- a/mindinsight/mindconverter/README.md
+++ b/mindinsight/mindconverter/README.md
@@ -394,6 +394,14 @@ Q2. Can MindConverter run on ARM platform?
 Q3. Why did I get message of `Error detail: [NodeInputMissing] ...` when converting PyTorch model?
 > Answer: For PyTorch model, if operations in `torch.nn.functional.xxx`, `torch.xxx`, `torch.Tensor.xxx` were used, node parsing could be failed. It's better to replace those operations with `torch.nn.xxx`.
 
+Q4. Why does the conversion process take a lot of time (more than 10 minutes), but the model is not so large?
+> Answer: When converting, MindConverter needs to use protobuf to deserialize the model file. Please make sure that the protobuf installed in Python environment is implemented by C++ backend. The validation method is as follows. If the output is "python", you need to install Python protobuf implemented by C++ (download the protobuf source code, enter the "python" subdirectory in the source code, and use `python setup.py install --cpp_implementation` to install). If the output is "cpp" and the conversion process still takes a long time, please add environment variable `export PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=cpp` before conversion.
+
+```python
+from google.protobuf.internal import api_implementation
+print(api_implementation.Type())
+```
+
 ## Appendix
 
 ### TensorFlow Pb model exporting
diff --git a/mindinsight/mindconverter/README_CN.md b/mindinsight/mindconverter/README_CN.md
index c5375e3c..14cda0d1 100644
--- a/mindinsight/mindconverter/README_CN.md
+++ b/mindinsight/mindconverter/README_CN.md
@@ -407,6 +407,15 @@ Q3. PyTorch模型转换时为什么提示`Error detail: [NodeInputMissing] ...`?
 
 > 答：对于PyTorch模型，若网络中存在`torch.nn.functional.xxx`, `torch.xxx`, `torch.Tensor.xxx`层算子，可能存在节点解析失败的情况，需要用户手动替换为torch.nn层算子。
 
+Q4. 为什么使用MindConverter进行模型转换需要很长时间（超过十分钟），而模型并不大？
+
+> 答：MindConverter进行转换时，需要使用Protobuf对模型文件进行反序列化，请确保Python环境中安装的Protobuf采用C++后端实现，检查方法如下，若输出为python，则需要安装采用C++实现的Python Protobuf（下载Protobuf源码并进入源码中的python子目录，使用python setup.py install --cpp_implementation进行安装）；若输出为cpp，转换过程仍耗时较长，请在转换前使用添加环境变量`export PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=cpp`。
+
+```python
+from google.protobuf.internal import api_implementation
+print(api_implementation.Type())
+```
+
 ## 附录
 
 ### TensorFlow Pb模型导出