Project Skylar is my attempt at making a digital assistant that can compete with the likes of Google Home and Amazon’s Alexa. I’ve been working on this project on and off since 2013, back then my goal was to create something similar to that of J.A.R.V.I.S. from the movie Iron Man. I wanted my own digital butler, one that could schedule appointments for me, manage the devices in my home, and answer questions when asked. This was quite an ambitious project that I had taken under my belt, I was just starting high school at the time and I only knew how to program basic things in Python.
Fast forward 6 years, now I’m a sophomore in college with a lot more programming experience. I’ve been working on this project off and on since then, I’ve gone through about 3 previous iterations, constantly changing the entire structure of the project each time.
The current iteration is the 4th, I call it Mark 4 (the Tony Stark in me couldn’t help it). Mark 4 is written in Python 3, my favorite programming language. There are 3 main components, the Core, Language Engine, and Networking. The core handles all of the inter-process communication and information handling that Skylar needs to function. The Language Engine is what turns the inputted utterance from the user into a set of actionable items that Skylar can perform. The final piece is networking, it handles all of the communication between Skylar and the internet. While these aren’t the only components Skylar needs to function, I feel as though they are the most important ones. As of right now, I’ve only fully implemented the Language Engine and the core, networking is still a work in progress.
There are quite a few features and abilities I want Skylar to have. I want Skylar to have a modular system structure, facial recognition, gesture recognition, P2P networking, hardware integration, and so much more. I plan on making regular blog posts documenting my progress with Skylar and keeping readers updated on any major breakthroughs I make. If I feel like the project is really going somewhere I may even consider releasing code as well. That’s it for this introduction, stay tuned for the next update where we will be discussing the Language Engine and Intent Classification!
Thanks for reading!