Building a Basic Chatbot with Azure OpenAI

Overview

In this turorial, we'll build a simple chatbot that uses Azure OpenAI to generate responses to user queries.

To create a basic chatbot, we need to set up a language model resource that enables conversation capabilities. In this tutorial, we will:

1. Set up the Azure OpenAI resource using the Azure AI Foundry portal.
2. Retrieve the API key needed to connect the resource to your chatbot application.

Once the API key is configured in your code, you will be able to integrate the language model into your chatbot and enable it to generate responses.

By the end of this tutorial, you'll have a working chatbot that can generate responses using the Azure OpenAI model.

Signing In and Setting Up Your Azure AI Foundry Workspace

Signing In to Azure AI Foundry

1. Open the Azure AI Foundry page in your web browser.
2. Login to your Azure account. If you don't have an account, you can sign up.
   

Setting Up Your Azure AI Foundry Workspace

1. Select + Create project to create a new project.
   
2. Perform the following tasks:
   • Enter Project name. It must be a unique value.
   • Select Hub you'd like to use (create a new one if needed).
3. Select Create.
   

    

Setting Up the Azure OpenAI Resource in Azure AI Foundry

In this step, you'll learn how to set up the Azure OpenAI resource in Azure AI Foundry. Azure OpenAI is a pre-trained language model that can generate responses to user queries. We'll be using it in our chatbot.

1. Select Models + endpoints from the left side menu.
   
2. On this page, you can deploy language models and set up Azure AI resources. In this step, we will deploy the Azure OpenAI GPT-4 language model. Select + Deploy model.
3. Select Deploy base model.
   
4. In this tutorial, we will deploy the GPT-4o model. Select GPT-4o.
5. Select Confirm.
   
6. Select Deploy.
   
7. The model will be deployed. Once the deployment is complete, you will see the model listed on the Models + endpoints page.
   
8. Now that the model is deployed, you can retrieve the API key needed to connect the model to your chatbot application. Select the model you deployed on the Models + endpoints page.
   `
9. On the model details page, you can view information about the model, including the API key. We will come back this page later to add the required information into the environment variables.

Setting Up the Project and Install the Libraries

Now, you will create a folder to work in and set up a virtual environment to develop a program.

Creating a Folder to Work Inside It

1. Open a terminal window and type the following command to create a folder named basic-chatbot in the default path.
   
   

   mkdir basic-chatbot

    

2. Type the following command inside your terminal to navigate to the basic-chatbot folder you created.
   
   

   cd basic-chatbot

Creating a Virtual Environment

1. Type the following command inside your terminal to create a virtual environment named .venv.
   
   

   python -m venv .venv

    

2. Type the following command inside your terminal to activate the virtual environment.
   
   

   .venv\Scripts\activate.bat

    

NOTE
If it worked, you should see (.venv) before the command prompt.

Installing the Required Packages

1. Type the following commands inside your terminal to install the required packages.
   • openai: A Python library that provides integration with the Azure OpenAI API.
   • python-dotenv: A Python library for managing environment variables stored in an .env file.
     
     

     pip install openai python-dotenv

      

Setting up the Project in Visual Studio Code

To create a basic chatbot program, you will need two files:

1. example.py: This file will contain the code to interact with Azure resources.
2. .env: This file will store the Azure credentials and configuration details.

NOTE

Purpose of the .env File

The .env file is essential for storing the Azure information required to connect and use the resources you created.
By keeping the Azure credentials in the .env file, you can ensure a secure and organized way to manage sensitive information.

Setting Up example.py File

1. Open Visual Studio Code.
2. Select File from the menu bar.
3. Select Open Folder.
4. Select the basic-chatbot folder that you created, which is located at C:\Users\yourUserName\basic-chatbot.
5. In the left pane of Visual Studio Code, right-click and select New File to create a new file named example.py. 
6. Add the following code to the example.py file to import the required libraries.
   
   

   from openai import AzureOpenAI
   from dotenv import load_dotenv
   import os
   
   # Load environment variables from the .env file
   load_dotenv()
   
   # Retrieve environment variables
   AZURE_OPENAI_ENDPOINT = os.getenv("AZURE_OPENAI_ENDPOINT")
   AZURE_OPENAI_API_KEY = os.getenv("AZURE_OPENAI_API_KEY")
   AZURE_OPENAI_MODEL_NAME = os.getenv("AZURE_OPENAI_MODEL_NAME")
   AZURE_OPENAI_CHAT_DEPLOYMENT_NAME = os.getenv("AZURE_OPENAI_CHAT_DEPLOYMENT_NAME")
   AZURE_OPENAI_API_VERSION = os.getenv("AZURE_OPENAI_API_VERSION")
   
   # Initialize Azure OpenAI client
   client = AzureOpenAI(
       api_key=AZURE_OPENAI_API_KEY,
       api_version=AZURE_OPENAI_API_VERSION,
       base_url=f"{AZURE_OPENAI_ENDPOINT}/openai/deployments/{AZURE_OPENAI_CHAT_DEPLOYMENT_NAME}"
   )
   
   print("Chatbot: Hello! How can I assist you today? Type 'exit' to end the conversation.")
   while True:
       user_input = input("You: ")
       if user_input.lower() == "exit":
           print("Chatbot: Ending the conversation. Have a great day!")
           break
       response = client.chat.completions.create(
           model=AZURE_OPENAI_MODEL_NAME,
           messages=[
               {"role": "system", "content": "You are a helpful assistant."},
               {"role": "user", "content": user_input}
           ],
           max_tokens=200
       )
       print("Chatbot:", response.choices[0].message.content.strip())

Setting Up .env File

To set up your development environment, we will create a .env file and store the necessary credentials directly.

NOTE
Complete folder structure:

└── YourUserName
.         └── basic-chatbot 
.         ├── example.py 
.         └── .env

1. In the left pane of Visual Studio Code, right-click and select New File to create a new file named .env.
2. Add the following code to the .env file to include your Azure information.
   
   

   AZURE_OPENAI_API_KEY=your_azure_openai_api_key
   AZURE_OPENAI_ENDPOINT=https://your_azure_openai_endpoint
   AZURE_OPENAI_MODEL_NAME=your_model_name
   AZURE_OPENAI_CHAT_DEPLOYMENT_NAME=your_deployment_name
   AZURE_OPENAI_API_VERSION=your_api_version

Retrieving Environment Variables from Azure AI Foundry

Now, you will retrieve the required information from Azure AI Foundry and update the .env file.

1. Go to the Models + endpoints page and select your deployed model.
   
2. On the Model Details page, copy the following information in to the .env file.:
   • AZURE_OPENAI_API_KEY
   • AZURE_OPENAI_ENDPOINT
   • AZURE_OPENAI_MODEL_NAME
   • AZURE_OPENAI_CHAT_DEPLOYMENT_NAME
     
     
3. Paste this information into the .env file in the respective placeholders.

Running the Chatbot Program

1. Type the following command inside your terminal to run the program and see if it can answer questions.
   
   

   python example.py

    

2. Interact with the chatbot by typing your questions or messages. The chatbot will generate responses based on the Azure OpenAI model you deployed.
   
   

NOTE

You can find the full example of this chatbot, including the code and .env template, in my GitHub repository: GitHub Repository