Skip to main content

Here's how to make a DIY Amazon Echo

Here's how to make a DIY Amazon Echo

Share this story

If you buy something from a Verge link, Vox Media may earn a commission. See our ethics statement.

Amazon’s Echo is great. It’s a speaker for music, a digital assistant with Amazon’s excellent Alexa voice control, and a useful smart home controller. But why buy an Echo from Amazon when you could build one yourself? Here’s how:


To put together your own Echo, you’ll need a few supplies. The most important thing is a Matrix Creator or Matrix Voice. They’re development platforms that connect to a Raspberry Pi and are designed to help makers test new hardware applications at a low cost. More importantly for us, they feature an eight-microphone array, which is perfect for creating our DIY Alexa. Any microphone will technically work, but the multiple microphone array is the closest you’ll get to Amazon’s own system on a budget.

Next, we’ll need the aforementioned Raspberry Pi. If the Matrix is the ears of our DIY Echo, the Pi is the brains. It’ll supply the processing power to run Amazon’s Alexa Voice Service, the audio output to connect a speaker, and Wi-Fi or Ethernet to connect to Amazon’s servers to get the answers to whatever you ask your Alexa. A Raspberry Pi 3 will have onboard Wi-Fi, but if you have an older one, you’ll need to pick up a USB adapter or plug in via Ethernet. You’ll also need a keyboard, mouse, and display to hook up to your Pi for the setup process.

Optionally, you’ll want a USB battery pack, to allow for your DIY Echo to work away from a wall outlet, a speaker for improved audio output, and some cardboard and tape to make a case.

Putting it together

We’ll be working off the official Alexa GitHub repository’s guide to running the Alexa Voice Service on a Raspberry Pi so you’ll want to follow along here. Additionally, we’ll be using some modifications from Matrix to take advantage of the more advanced mics the Creator or Voice have to offer.

First off, you’ll want to hook up your Raspberry Pi to your screen, keyboard, and mouse, and then plug in your Matrix Creator. Following the guide from the Alexa GitHub, you’ll need to create an Amazon developer account, and register a device — in this case, our Raspberry Pi — with a device and security profile.

Next, you’ll need to install the Amazon Voice Service sample app from the GitHub page. Once you’ve gotten the installer ready, continue to follow the instructions to input the security profile information from the earlier step, and then begin the installation. Let it run; it’ll take roughly 20–30 minutes to finish.

Once that’s done, flip over to the Matrix guide: you’ll need to configure the Pi to use the Matrix microphones, and install the Matrix software.

Then, all you need to do is follow the final step from the Alexa GitHub guide, and start the three software processes you need to access the Alexa voice service. Unfortunately, you’ll have to do that every time you boot up the Pi to get the Alexa services running. While it is possible to do some more work in the terminal to get all this to launch at startup so the keyboard mouse and screen aren't required every time, that's beyond the scope of this tutorial.

Once that’s up and running, congratulations! You’ll have your very own Echo. But to really finish things off, you’ll want to cut out a tube of cardboard, hook up a battery pack and speaker, disconnect the mouse, screen, and keyboard, and tape everything up together in a roughly cylindrical package for the true faux-Echo experience.