Connect IoT devices to Azure AI Speech using Azure Functions

  • Module
  • 13 Units
Intermediate
AI Edge Engineer
Azure IoT Hub
Azure Functions

Create and deploy a cloud-hosted HTTP-triggered Azure Function that connects an MXChip IoT DevKit to Azure AI Speech. The device records a tourist's spoken source language and posts the audio to the function. The function translates the speech into English by default, sends the translated text back through IoT Hub cloud-to-device messaging, and the IoT DevKit displays it.

Learning objectives

In this module, you will:

  • Configure an MXChip IoT DevKit with an IoT Hub
  • Integrate Azure AI Speech translation into a cloud-hosted Azure Function
  • Deploy an Azure Function app
  • Test your Azure Function app by sending translated text to the IoT DevKit through IoT Hub cloud-to-device messaging

"Produced in partnership with the University of Oxford – Ajit Jaokar, Artificial Intelligence: Cloud and Edge Implementations course."

Prerequisites

  • Familiarity with Azure
  • Basic understanding of IoT
  • Basic knowledge of Azure Functions
  • Basic knowledge of Azure AI services
  • A MXChip IoT DevKit board with a Micro-USB cable
  • A computer running an operating system supported by current Visual Studio Code and Azure Functions tooling; use the older OS versions only when reproducing the archived legacy sample with matching legacy tools.
  • An active Azure subscription with capacity to create or reuse an IoT Hub. You can use the free tier only if the subscription doesn't already have a free-tier IoT Hub.
  • Visual Studio Code installed on your computer
  • Arduino IDE installed on your computer
  • A supported .NET SDK installed on your computer, such as .NET 8, if you migrate the Functions project to Azure Functions v4. The legacy .NET Core 2.1 SDK is only for archival reproduction of the old sample and isn't supported for current Azure deployment.
Start