VEDA logo
Menu
日本語

Audio Input

Click this button to see an example of GLSL using audio inputs.

VEDA supports audio input.
When you set audio: true in Settings, VEDA enables following uniform variables:

  • float volume
  • sampler2D samples
  • sampler2D spectrum

sampler2D samples stores the most recent 256 frames from the audio input.
This is useful for drawing waveforms.

sampler2D spectrum stores the FFT result.
This is useful to draw the volume of specific frequency band, such as spectrum visualizer.

float volume is the average of all the frequency bands in spectrum.

See examples for more details.

Example

This is the code running on this page (PC only).

/*{ "audio": true }*/
precision mediump float;
uniform float time;
uniform vec2 resolution;
uniform sampler2D spectrum;
uniform sampler2D backbuffer;

float random(in vec2 p) { ... }
float noise (in vec2 st) { ... }
float fbm(in vec2 p) { ... }
vec2 rotate(vec2 p, float t) { ... }

void main() {
  vec2 p = (gl_FragCoord.xy * 2. - resolution) / min(resolution.x, resolution.y);
  vec2 uv = gl_FragCoord.xy / resolution;

  float a = atan(p.y, p.x);
  float l = length(p);
  p += fbm(vec2(p + l * 20. + a * sin(time * .3) * 10.)) * .2;

  p = vec2(length(p - .1));
  p -= time * .3;

  gl_FragColor = vec4(
    texture2D(spectrum, mod(p * .39 - time * .19, .7)).r,
    texture2D(spectrum, mod(p * .31 - time * .13, .54)).r,
    texture2D(spectrum, mod(p * .37 - time * .17, .6)).r,
    1.
  ) + texture2D(backbuffer, uv)*.8;
}