This is a tutorial to show how one can integrate Rasa chatbots with Unity.
Currently no work on chatbot has been done except implementing the custom connector. Chatbot is trained on the Mood bot data that is created after rasa init
command.
First, clone this repo using git clone https://github.com/retrogeek46/Rasa-Unity
and checkout to v0.2.0.
- Change directory into Rasa/ and create a virtual environment using
python -m venv venv
- Activate the virtual environment
Bash on Windows :source venv/Scripts/activate
CMD :venv\Scripts\activate
Linux :source venv/bin/activate
- Install Rasa into the virtual environment (elevated prompt recommended for creating spacy link)
pip install rasa[spacy]
python -m spacy download en_core_web_md
python -m spacy link en_core_web_md en
- Run rasa server using
rasa run
- The Unity project was made using 2019.3 hence is the recommended version.
- Open Unity Hub and choose "Add" from top right corner.
- Select the Unity project folder located in Rasa-Unity/Unity/Rasa.
- Press play after running the Rasa server.
- Type your message and the bot should respond.
WIP, the source code is commented though.
- Clicking anywhere sends empty requests
Add scroll supportresize chat bubbles based on size of content (text, image)- Render buttons and quick replies
- Add support for custom data
- Render messages in order, wait for image loads
- Add docs
- Add delay between messages
- Add animations before showing messages
- Add voice support