Do you ever dream of living in smart-home like that of Iron Man and building a smartass robot assistant that would cater to all your needs without you even having to press a single button? We all secretly dream of living in an alternate world where super-heroes battle the evil villains using their super-powers, but let’s be honest, it’s all fiction and it has no place in the real world. But what if there was a way you could steal away a part of the fiction and bring it into the real world? Seems too good to be right? But what if we told you that Facebook founder Mark Zuckerberg has found a way to turn your this ‘fictional-dream’ into a reality?
Almost a year ago, Zuckerberg had announced that he would develop an AI that would help him run his home and help him with work. “My personal challenge for 2016 is to build a simple AI to run my home and help me with my work. You can think of it kind of like Jarvis in Iron Man,” he had written in his Facebook post. And almost a year later, he has successfully developed Jarvis, a smart-home assistant that not only follows simple commands like switching on the lights and controls temperature at his home but also plays songs based on his mood. Pretty amazing right? And now even you can build your very own Jarvis. Here’s how…
Step 1: Connect your home
The first and most important step in the process of building a smart-home AI is connecting your home on a common platform. As Zuckerberg explained that before working on building an AI, he first had to write codes to connect all the systems in his house, which speak different languages and run of different protocols. “I had to reverse engineer APIs for some of these to even get to the point where I could issue a command from my computer to turn the lights on or get a song to play,” he shared in a blog post.
ALSO READ: WTF is Artificial Intelligence!
Step 2: Teach your home to communicate with your computer
After inter-connecting your home devices, the next step is to connect your home to your computer. Zuckerberg explained that he did this in two steps. In the first step, he connected his system to his home such that they could connect communicate using text messages. In the next step, he added the ability to translate his speech into text for the system to read.
He started off with simple words like “bedroom” and “lights” and gradually added synonyms and context. “The more context an AI has, the better it can handle open-ended requests. At this point, I mostly just ask Jarvis to “play me some music” and by looking at my past listening patterns, it mostly nails something I’d want to hear.”
Step 3: Vision and face recognition
Face recognition is particularly useful when your friends are at your door and your AI needs to determine whether to let them in. To do this he installed a few cameras at his door that would capture images from all angles. He then built a server that continuously watched the cameras and ran a two-step process: ‘first, it runs face detection to see if any person has come into view, and second, if it finds a face, then it runs face recognition to identify who the person is.’ Once the server identifies the person, it checks a list to confirm if he’s expecting the person, and if he is then it will let the people in and inform him that they’re here.
Step 4: Connect your system to the messenger bot
For an AI to be truly fully functional, it’s important that you’re able to communicate with it from anywhere. For this to happen, it’s necessary that the communication should happen via phones. Zuckerberg used Messenger bot for this. “I started off building a Messenger bot to communicate with Jarvis.”
Step 5: Voice and Speech Recognition
Voice-based communication is as important as the text-based conversation. To enable voice for Jarvis, he built a dedicated Jarvis app that continuously listens to what he’s saying. He first built the iOS version of the Jarvis app and now he’s planning to develop the Android version.
With all the components in place, you are ready to rock on with your very own home-made Jarvis.