What Is That?

ROLE: Creative Technologist
DATE: February 2019
TOOLS: AIY Voice Kit, Tensorflow & Python
Remaking the AIY Voice Kit

The objective of this project is adding a Pi Camera to the AIY Voice Kit in order to let machines describe what they see. During the project I had a few questions for the AIY Kit Community. One Googler in particular (thanks Dmitriy!), took an interest in the project and asked me to publish on

The Challenge

Initially, I found a good tutorial by Leo White: "Adding Vision to the Voice Kit" which helped with both the build and code strategy. But his blog post is from March 2018 and >no longer compatible with the current AIY Kits. So I rewrote the code and enhanced the UX. To add a bit more entertainment value, I built in a few opportunities for personalization and customization as well.


In the past 4 months the project has received 1,404 views and 20 "respect" votes. It was only the second project I've published to and was a fun experience. Beyond the Hackster audience - I'm now using this techology in a commercial application. As they say, "necessity is the mother of invention" - and I'm now working on a fresh multi-modal (Voice/Vision/ML) experience to solve a common problem in modern kitchens. Stay tuned!

Adding the Camera Adding the Camera

Attaching the Speaker Attaching the Speaker

Adding an LED Adding an LED

Positioning the LED Positioning the LED

Finished Build Finished Build

Random Responses Make Better VUI/UX Using Randomization to Make Dialog Sound More Natural