using an ESP32 and OLED display together with an FFT to display 8-octaves of audio in the frequency domain
Switch branches/tags
Nothing to show
Clone or download
Latest commit 9e498a7 Oct 5, 2018
Type Name Latest commit message Commit time
Failed to load latest commit information.
ESP32_Spectrum_Display_03.ino Timing Updates May 17, 2018
ESP32_Stereo_Spectrum_Display_01.ino Timing Updates May 17, 2018
ILI9341 Version ILI9341 version by Tobozo Dec 21, 2017
Licence.txt Add files via upload Oct 5, 2018 Update Dec 20, 2017
Stereo_Input_Wiring.jpg Thicker Lines Mar 17, 2018
Wiring Diagram-ESP32a.JPG Wiring Diagram Dec 17, 2017
font.h Add files via upload Dec 17, 2017


Using an ESP32 and OLED display together with an FFT to display audio as 8-octaves in the frequency domain.

An improved version solely for the ESP32 with its significantly faster ADC to enable an increase in the resolution of the frequency domain by a factor of 4, whilst also giving an improved dynamic range. Code improvements have also been made to speed up the processing together with a shorter code length.

You must place the font.h file in the same location as the sketch. The Font file is needed to provide a font that can display enough characters across the screen to enable each band to be clearly denoted. The bands are:

125Hz 250Hz 500Hz 1KHz 2Khz 4KHz 8Khz 16kHz

Please note, if you use a microphone - speaker test environment it is highly likely (unless of very high bandwidth and quality) that both the microphone and speakers used have insufficient quality to ensure you are getting a flat frequency response for the Fast Fourier Transform to convert and display. The only way to ensure you see the correct result is to feed approximately 50mV to 100mV pk-pk of audio directly into the ADC port with no DC offset. This is not a fault with the software or display or FFT - it’s the laws of physics!