Securing access to Azure services

Think Different
2 min readNov 25, 2023

Securing access to Azure services, including Azure AI services like Speech, involves managing and protecting the authentication credentials (such as subscription keys or service principal credentials). Here are steps to securely handle these credentials in an Azure environment:

1. Azure Managed Identity (Recommended for Azure Functions):

- If your application is running in Azure, consider using Azure Managed Identity.

- Enable Managed Identity for your Azure Function in the Azure Portal.

- Grant the necessary permissions (like access to Azure Speech service) to the Managed Identity.

2. Azure Key Vault:

- Azure Key Vault is a secure way to store and manage sensitive information, such as API keys and secrets.

- Create a Key Vault in the Azure Portal.

- Store your Speech API key or other sensitive information securely in Azure Key Vault.

- Grant necessary permissions to your Azure Function to access the Key Vault.

3. Environment Variables:

- If you need to use environment variables, ensure they are stored securely.

- In Azure Functions, you can use the Azure Functions Application Settings to store environment variables securely.

- Avoid hardcoding sensitive information in your code.

4. Managed Identities for Azure Resources (MI for Azure Resources):

- Enable Managed Identities for your Azure Function App.

- Grant necessary permissions to the Managed Identity (for example, access to the Speech service).

5. Role-Based Access Control (RBAC):

- Use RBAC to control access to resources.

- Assign roles to your Azure Function’s Managed Identity based on the principle of least privilege.

Example: Using Azure Key Vault in Azure Functions (Python):

1. Configure Key Vault Reference in `local.settings.json`:



“IsEncrypted”: false,

“Values”: {

“AzureWebJobsStorage”: “your_storage_connection_string”,



“Host”: {

“LocalHttpPort”: 7071,

“CORS”: “*”


“ManagedDependency”: {

“Enabled”: true




Replace `”your_storage_connection_string”` with your actual storage connection string.

2. Reference Key Vault Secrets in Python Code:


import os

from azure.identity import DefaultAzureCredential

from azure.keyvault.secrets import SecretClient

key_vault_uri = “"

secret_name = “your-secret-name”

credential = DefaultAzureCredential()

secret_client = SecretClient(vault_url=key_vault_uri, credential=credential)

secret_value = secret_client.get_secret(secret_name).value


By using Azure Key Vault or Managed Identity, you enhance the security of your application by centralizing and securing your secrets, reducing the risk of exposure. Ensure that your application adheres to Azure security best practices and follows the principle of least privilege.



Think Different

My name is Dhiraj Patra a Software architect for AI, ML, IoT microservices cloud applications. Love to learn and share.