A spectrum analyzer is a great way to create exciting visuals that pulse in time with music. [pyrograf] wanted a big one as a display piece, so set about whipping up something of their very own.
An ESP32 microcontroller serves as the heart of the build, with its high clock rate and dual cores making it a highly capable choice for the job. Audio from a microphone is amplified and pumped into the ESP32’s analog input. Core 0 on the ESP32 then runs a Fast Fourier Transform on the input audio in order to determine the energy in each frequency band. The results of this FFT are then passed to Core 1, which is used to calculate the required animations and pipe them out to a series of WS2812B LEDs.
Where this build really shines, though, is in the actual construction. Big chunks of acrylic serve as diffusers for the LEDs which light up each segment of the spectrum display. Combine the big pixel size with a nice smooth 30 Hz refresh rate on the LEDs, and the result is a rather large spectrum analyzer that really does look the business.
We’ve seen some similar builds over the years, too. Video after the break.
2 thoughts on “Big Audio Visualizer Pumps With The Music”
Nice music for once but a test tone sweep or something would be nice. I can’t tell is it’s live or not. Most importantly if the time/freq base is linear or octave.
Thank you for sharing my project :) I answered question in YT comment. I used octaves, some more information on github. There is python script to generate bins for FFT.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)