Inspiration
We are part of a performing arts group on campus called Photonix that utilizes glowsticks, LEDs, and other sources of light to express our art. We wanted to create something that could use simple lights to do something pretty awesome.
What it does
The project (LEDsense) analyzes an inputted audio file (.mp3 and .wav are both consistently reliable) and automatically determines a light pattern that allows LEDs to flash/fade in accordance with the music.
How we built it
LEDsense consists of a software and a hardware portion. The software analyzes a valid audio input file and calculates the timestamps within the audio with the largest changes in frequency. This allows us to consistently determine unique and reliable light patterns for LEDs to follow that match up with the audio. The software outputs a .ino file that can be directly compiled and uploaded onto an Arduino/Microcontroller with LED lights.
The hardware uses all sorts of materials we found throughout the hackathon. The project primarily uses four "hackerboards" and microUSB-to-USB cables provided to us by MHacks. After uploading the code onto the microcontrollers, we used battery packs to make them portable. With some creativity and lots of tape, we advanced the design and improved light refraction using plastic bottles and trashbags gathered throughout the event.
Challenges we ran into
The actual process of turning audio into usable data proved to be pretty confusing and new to us. Also, as this was our first hackathon, we didn't know what to expect and it took a little bit to get things moving. After that, the challenge was how long we could manage to stay awake.
Accomplishments that we're proud of
Although crude, we never really expected to get this far into the project. We've successfully combined an advanced light mapping algorithm with equipment that retails at USD100+ with random material we found and had on hand. Pretty neat!
What we learned
This was our first hackathon and our first time working with audio/music as well as microcontrollers. The whole project was pretty much a hands-on learning experience!
What's next for LEDsense
We hope to continue improving the algorithm and light patterns. We can already customize the sensitivity and frequency of light changes by manipulating the threshold and sample rate, but there are many other factors and analyzable variables that can go into improving the algorithm. Clearly our hardware is still pretty crude, so we definitely hope to improve on that front!
Log in or sign up for Devpost to join the conversation.