Artificial Intelligence Yourself – Google Voice Kit

Google voice kit made by you with google is an experiment made by google and let to enthusiasts to improve Human Interface Interactions. 

Communication over the years evolved from being person to person -> person to person through machines -> person to intelligent machines. Over this journey user interface and experience remained the same. Now that companies are leapfrogging with technology advancements, VUI ( Voice User Interface) has become feasible at a reduced cost. 

We talk about robots – human intelligence in machines. But to have a seamless user experience those machines have to adopt to new user interface apart from responding to touch and RF signals. And voice is unanimous choice to enhance the human interaction with machines.  

What does AIY voice kit provide?

A simple way to understand how google home or amazon echo are designed and built by prototyping similar kind of devices. Also, more flexibility to try new use cases apart from simply assembling the hardware and loading the software on to it. 

Google Voice Kit
Prototyping with Voice Kit

Enriched VUI

Unlike the usual echo or home which can be invoked by saying alexa or hey google respectively – Voice kit can also be invoked with custom phrases with a limitation of holding the button. Direct interaction capability would eliminate middle mans like echo and home. 

Hardware requirements – 

Voice kit has both speakers and microphone. However you will need to buy raspberry Pi 3 separately.

Raspberry is responsible for – 

1) Internet connectivity

2) Connecting to google cloud

3) Transporting the request and response off the voice kit

4) Software runs on the pi

5) Any further tweak to programming is done at pi

Other items that you may want to buy – Power supply to PI and SD card


Download the latest version of aiyprojects image from AIY website. This has to be burnt to the SD card that will go with raspberry pi. Use Etcher for the same.

Introduction to new google platforms and tools – 

While you run through all the chapters you will realize you are getting exposed to some of the below – 

Google cloud platform – Shorty, GCP is a suite of cloud computing services offered by Google. They offer seamless server-less computing similar to AWS. While you do AIY voice kit you will create a GCP project and create credentials to use it for various API’s.

Google machine learning API’s – Google machine learning API’s do the heavy lifting operation of performing at scale. They have both pre-trained models and a service to train your own models. You will try to use speech recognition API as part of prototyping.

TensorFlow – TensorFlow is an open source deep learning platform by google. With the launch of TensorFlow Lite for mobile and embedded devices you can try out some models on raspberry pi 3 itself. This adds offline intelligence to the machines.

Using python

A simple easy to learn object oriented programming language. You will have a chance to play around some simple python programs as you implement and improve use cases. Also play around some of the API’s that AIY has provided.

Finally, Shutdown properly

If you leave the setup as such there is a tendency of the SD card to get corrupted. Remember there are lots of ways to add a shutdown button to your kit by googling around the solution.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.