Stephen Hawking’s improved speech software is now free to download

In January 2012 Intel announced they were looking into improving the efficiency of Stephen Hawking’s access to his communication aid. At the time, an infrared sensor mounted to his glasses detected small pulses of movement in his cheek. This enabled him to write at around 15 words per minute using a 20-year old software packaged called EZ Keys. EZ Keys is a functional, if not pretty, switch-accessible keyboard with prediction. Stephen wanted a more efficient system and one which made fewer mistakes. The power of movement in his cheek was also beginning to fade which affected the reliability of the sensor. Intel began modifying the software while experimenting with high-definition facial recognition cameras, eye gaze sensors and brain wave devices.

EZ Keys
Custom-made Intel Tablet running EZ Keys in 2011. Intel Free Press.

At the end of 2014 Intel announced that they had developed software which doubles Stephen’s input speed and provides faster access to common tasks (such as a simple ‘back’ button). It makes use of improved prediction technology from London startup SwiftKey.

“My old system was over 20 years old and I was finding it very difficult to communicate effectively. This new system is life changing for me and I hope will serve me well for the next 20 years.”

The new open-source software package, Assistive Context Aware Toolkit (ACAT), has today been posted online for people to freely download, thus making it available for the 3 million people worldwide whose lives are affected by quadriplegia, motor neurone disease and other physical disabilities.

Stephen is still using his cheek to access the machine as the facial recognition system and brain waves were found to be inconsistent so far.

Stephen Hawking
Stephen Hawking at Gonville & Caius College, Cambridge. Lwp Kommunikáció/

Jonathan Wood, Stephen’s graduate assistant who has piles of experimental communication hardware and interfaces in his office next to Stephen’s, is hopeful that a high-resolution camera will provide the solution to existing inefficiencies and errors made through the cheek switch.

“The idea is to have a camera pointed at Stephen’s face to pick up not just his cheek movements but other facial movements,” says Wood. “He could move his jaw sideways, up and down, and drive a mouse and even potentially drive his wheelchair. These are cool ideas but they won’t be coming to completion any time soon.”

You can download ACAT from I’ll post a review of the software at a later date.


Featured Image: Jim Campbell/Aero-News Network

Leave a comment

Your email address will not be published. Required fields are marked *

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