The making of Voice Commander

From idea to product

"Working in the video game industry as a sound designer, I watched colleagues waste their time conducting repetitive actions of all kinds--tedious computer tasks that added hours of unnecessary work to their day. These tasks constituted a maze preventing them to get from point A to point B. Being a perfectionist, I enjoyed working extra hours to please my supervisor, even when it meant sacrificing my personal time and wellbeing. Work always came before relationships. But as I approached my 30s, my priorities started to evolve. I wanted to live a balanced, full and healthy life, but a lack of time was holding me back from achieving this upgrade. Being confident in my execution skills, I searched for ways to dramatically streamline my workflow in order to recapture precious time."

"I developed new software tools connected to hardware controllers. This creative project enabled me to iterate over functions, handle recursive actions, and navigate effortlessly through my work day. I was trying to automate my workflow and expand the amount of inbound commands at my workstation using additional hardware controllers. At one point my setup involved pad controllers on the left side of my keyboard, a circular motion wheel at the right of my mouse and pedals for my feet. This turned out to be overkill! I could not process my thoughts and software commands using all these devices at once. It felt like conducting an orchestra and trying to drum at the same time. All this hardware was not accessible from other rooms of the office and was a pain to move to other workstations. Plus, learning how to use all the equipment and execute commands ended up making me work slower, not faster for the first few weeks. How could I get rid of this complex clutter?

I needed something hands-free that I could access everywhere, something that felt natural to use so I wouldn’t have to invest time in getting the hang of it. A tool to understand my commands and process extensive operations quickly. That’s when it dawned on me: I needed a “ubiquitous virtual assistant” by my side that could replace hundreds of controllers." What technology could meet the bold requirements of remote communication between me and my computer, with fast processing, operate hands free, require no learning curve and no additional hardware? It’s almost like I wanted my computer to read my mind!

I started to learn about brain computer interfaces. Developing such a product at this early stage of the technology was very risky. It also required advanced hardware engineering for the cap or headband to record the electrical activity of the brain. A very poor chance I’d make that work. I needed a new strategy. What natural process comes after the birth of a thought? We often project a mental image; no way I could capture that. What comes next then? We often communicate an idea by putting it into words. Bingo!

Speech recognition is now available in our mobile apps to input text; could it also automate process via custom commands? Speech comes naturally to us, does not require special hardware to capture, and it could be completely remote if I could pair my software with my iPhone and send commands over the network. I looked up the available API’s and luckily their were plenty of options. Some of them included layers of deep learning that would improve the recognition accuracy over time. This sounded so promising, especially since Siri, a popular voice-activated assistant, had merged with desktop computers, and the tool Microsoft Garage now allowed some voice commands with Microsoft Word. Since nothing yet existed for graphic software and digital audio workstations. I saw an opportunity to innovate, and here I bring you my creation, Voice Commander. Well then, let’s get to work!"