{"id":5392,"date":"2023-02-23T18:47:22","date_gmt":"2023-02-23T17:47:22","guid":{"rendered":"https:\/\/www.pschatzmann.ch\/home\/?p=5392"},"modified":"2023-07-14T18:32:56","modified_gmt":"2023-07-14T16:32:56","slug":"arduino-audiotools-display-fft-result-on-a-led-matrix","status":"publish","type":"post","link":"https:\/\/www.pschatzmann.ch\/home\/2023\/02\/23\/arduino-audiotools-display-fft-result-on-a-led-matrix\/","title":{"rendered":"Arduino AudioTools: Display the FFT Result on a LED Matrix"},"content":{"rendered":"

I have added a simple output class to my Arduino Audio Tools<\/a> which can display the result of a FFT analysis on an LED Matrix. For my tests, I am using a 32 x 8 LED Matrix<\/strong>, but first we need to test the functionality.<\/p>\n

\"\"<\/p>\n

Determine the Wiring Logic<\/h3>\n

I am using the FastLED<\/a> library to update the LEDs. To simplify the processing I am providing the LEDOutput<\/a> class.<\/p>\n

#include \"AudioTools.h\"\n#include \"AudioLibs\/LEDOutput.h\"\n\n#define PIN_LEDS 22\n#define LED_X 32\n#define LED_Y 8\n\nLEDOutput led; \/\/ output to LED matrix\nint pos = 0;\n\nvoid setup() {\n  Serial.begin(115200);\n  AudioLogger::instance().begin(Serial, AudioLogger::Info);\n\n  \/\/ Setup LED matrix output\n  auto lcfg = led.defaultConfig();\n  lcfg.x = LED_X;\n  lcfg.y = LED_Y;\n  led.begin(lcfg);\n\n  \/\/ add LEDs\n  FastLED.addLeds<WS2812B, PIN_LEDS, GRB>(led.ledData(), led.ledCount());\n}\n\nvoid loop() { \n   led.led(pos++) = CRGB::Red;\n   FastLED.show();\n   delay(500);\n}\n<\/code><\/pre>\n
The sketch above demonstrates that the matrix uses a serpentine layout<\/strong>.<\/p>\n
Testing the Wiring Logic<\/h3>\n
We can use the is_serpentine_layout<\/strong> parameter to select the serpentine layout.<\/p>\n
#include \"AudioTools.h\"\n#include \"AudioLibs\/LEDOutput.h\"\n\n#define PIN_LEDS 22\n#define LED_X 32\n#define LED_Y 8\n\nLEDOutput led; \/\/ output to LED matrix\nint x = 0;\nint y = 7;\n\nvoid setup() {\n  Serial.begin(115200);\n  AudioLogger::instance().begin(Serial, AudioLogger::Info);\n\n  \/\/ Setup LED matrix output\n  auto lcfg = led.defaultConfig();\n  lcfg.x = LED_X;\n  lcfg.y = LED_Y;\n  lcfg.is_serpentine_layout = true\n  led.begin(lcfg);\n\n  \/\/ add LEDs\n  FastLED.addLeds<WS2812B, PIN_LEDS, GRB>(led.ledData(), led.ledCount());\n}\n\nvoid loop() { \n   led.ledXY(x++,y) = CRGB::Red;\n   FastLED.show();\n   delay(500);\n}\n<\/code><\/pre>\n
The sketch above should update the leds from the top row from left to right<\/strong>!<\/p>\n
The FFT Output<\/h3>\n
Finally here is the sketch that uses the AudioKit as input<\/strong> and displays the FFT result on the LED Matrix<\/strong>. If you missed my blog about FFT, I suggest that you have a look at it<\/a>.<\/p>\n
#include \"AudioTools.h\"\n#include \"AudioLibs\/AudioKit.h\"\n#include \"AudioLibs\/AudioRealFFT.h\" \/\/ or AudioKissFFT\n#include \"AudioLibs\/LEDOutput.h\"\n\n#define PIN_LEDS 22\n#define LED_X 32\n#define LED_Y 8\n\nAudioKitStream kit;  \/\/ Audio source\nAudioRealFFT fft; \/\/ or AudioKissFFT\nLEDOutput led(fft); \/\/ output to LED matrix\nStreamCopy copier(fft, kit);  \/\/ copy mic to fft\n\nvoid setup() {\n  Serial.begin(115200);\n  AudioLogger::instance().begin(Serial, AudioLogger::Info);\n\n  \/\/ setup Audiokit as input device\n  auto cfg = kit.defaultConfig(RX_MODE);\n  cfg.input_device = AUDIO_HAL_ADC_INPUT_LINE2;\n  kit.begin(cfg);\n\n  \/\/ Setup FFT output\n  auto tcfg = fft.defaultConfig();\n  tcfg.length = 1024; \n  tcfg.copyFrom(cfg);\n  fft.begin(tcfg);\n\n  \/\/ Setup LED matrix output\n  auto lcfg = led.defaultConfig();\n  lcfg.x = LED_X;\n  lcfg.y = LED_Y;\n  lcfg.is_serpentine_layout = true\n  lcfg.fft_group_bin = 3;\n  lcfg.fft_start_bin = 0;\n  lcfg.fft_max_magnitude = 40000;\n  led.begin(lcfg);\n\n  \/\/ add LEDs\n  FastLED.addLeds<WS2812B, PIN_LEDS, GRB>(led.ledData(), led.ledCount());\n}\n\nvoid loop() { \n  \/\/ update FFT\n  copier.copy();\n  \/\/ update LEDs\n  led.update();\n}\n<\/code><\/pre>\n
We use a fft length of 1024 which gives 512 frequency buckets. We support plenty of led configuration parameters<\/a>: When we output the data we put the buckets into groups of 3. The max magnitude parameter will drive how sensitive the display of the bars will be.<\/p>\n
You can define your own callback functions<\/strong> to define your custom update logic or for determining the colours.<\/p>\n
Here are the examples that can be found on Github:<\/p>\n