diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 0000000..c539e8c
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,18 @@
+FROM python:3.10.7
+
+RUN pip install --upgrade pip
+
+ADD . /app
+
+WORKDIR /app/
+
+RUN pip install .
+
+RUN pip install jupyter
+
+RUN apt update && apt-get -y install libsndfile-dev
+
+EXPOSE 8840
+
+ENTRYPOINT ["jupyter", "notebook", "--ip=0.0.0.0" , "--port=8840", "--no-browser", "--allow-root"]
+
diff --git a/README.md b/README.md
index 2d5a1e9..44e32ed 100644
--- a/README.md
+++ b/README.md
@@ -19,12 +19,24 @@ An unofficial PyTorch implementation of [VALL-E](https://valle-demo.github.io/),
 Since the trainer is based on [DeepSpeed](https://github.com/microsoft/DeepSpeed#requirements), you will need to have a GPU that DeepSpeed has developed and tested against, as well as a CUDA or ROCm compiler pre-installed to install this package.
 
 ### Install
+You can install vall-e using 3 methods:
+1. Docker
 
+```
+git clone --recurse-submodules https://github.com/enhuiz/vall-e.git
+cd vall-e
+docker build -t valle/python3.10 .
+docker run -it --gpus=all --net=host --ipc=host -v $(pwd):/app/ valle/python3.10
+
+```
+The last command will run the docker image using all the available GPUs, you can refer to the [documentation](https://docs.docker.com/config/containers/resource_constraints/#:~:text=an%20error%20occurs.-,GPU,-%F0%9F%94%97) to specify the GPUs you want to use. Also, it will use a volume that keeps all of the container's files shared between the container's `/app` and the current working directory of the server `$(pwd)`.
+
+1. Pip
 ```
 pip install git+https://github.com/enhuiz/vall-e
 ```
 
-Or you may clone by:
+1. Clone
 
 ```
 git clone --recurse-submodules https://github.com/enhuiz/vall-e.git