Merge branch 'main' into glossing/lfo-upgrades
This commit is contained in:
commit
12f9af227c
18 changed files with 699 additions and 146 deletions
|
|
@ -11,6 +11,9 @@ const PI = Math.PI;
|
|||
const TWO_PI = 2 * PI;
|
||||
const INVSR = 1 / sampleRate;
|
||||
|
||||
const timeToCoeff = (t) => 1 - Math.exp(-INVSR / t);
|
||||
const dbToLin = (db) => Math.pow(10, db / 20);
|
||||
|
||||
const clamp = (num, min, max) => Math.min(Math.max(num, min), max);
|
||||
const mod = (n, m) => ((n % m) + m) % m;
|
||||
const lerp = (a, b, n) => n * (b - a) + a;
|
||||
|
|
@ -1391,3 +1394,82 @@ class WavetableOscillatorProcessor extends AudioWorkletProcessor {
|
|||
}
|
||||
|
||||
registerProcessor('wavetable-oscillator-processor', WavetableOscillatorProcessor);
|
||||
|
||||
class TransientProcessor extends AudioWorkletProcessor {
|
||||
static get parameterDescriptors() {
|
||||
return [];
|
||||
}
|
||||
|
||||
constructor(options) {
|
||||
super();
|
||||
this.gainCoeff = timeToCoeff(0.2);
|
||||
this.avgGain = 1;
|
||||
let {
|
||||
attackTime = 0.003,
|
||||
sustainTime = 0.08,
|
||||
attack = 0,
|
||||
sustain = 0,
|
||||
sensitivity = 0.1,
|
||||
mix = 1,
|
||||
begin = 0,
|
||||
end = 0,
|
||||
} = options.processorOptions;
|
||||
attackTime = clamp(attackTime, 0.0005, 0.05);
|
||||
sustainTime = clamp(sustainTime, 0.01, 0.5);
|
||||
this.attackCoeff = timeToCoeff(attackTime);
|
||||
this.sustainCoeff = timeToCoeff(sustainTime);
|
||||
this.attackAmt = clamp(attack, -1, 1);
|
||||
this.sustainAmt = clamp(sustain, -1, 1);
|
||||
this.scaling = 0.5 + 5 * clamp(sensitivity, 0, 1);
|
||||
this.mix = clamp(mix, 0, 1);
|
||||
this.begin = begin;
|
||||
this.end = end;
|
||||
this.attackEnv = new Float32Array(2); // assume stereo
|
||||
this.sustainEnv = new Float32Array(2);
|
||||
}
|
||||
|
||||
process(inputs, outputs, _params) {
|
||||
const input = inputs[0];
|
||||
const output = outputs[0];
|
||||
if (currentTime >= this.end) {
|
||||
return false;
|
||||
}
|
||||
if (currentTime <= this.begin) {
|
||||
return true;
|
||||
}
|
||||
const channels = input.length;
|
||||
if (channels > this.attackEnv.length) {
|
||||
this.attackEnv = new Float32Array(channels);
|
||||
this.sustainEnv = new Float32Array(channels);
|
||||
}
|
||||
let avgGain = this.avgGain;
|
||||
for (let ch = 0; ch < channels; ch++) {
|
||||
let attEnv = this.attackEnv[ch];
|
||||
let susEnv = this.sustainEnv[ch];
|
||||
for (let n = 0; n < blockSize; n++) {
|
||||
const sample = input[ch][n];
|
||||
const x = Math.abs(sample);
|
||||
attEnv = lerp(attEnv, x, this.attackCoeff);
|
||||
susEnv = lerp(susEnv, x, this.sustainCoeff);
|
||||
const peakiness = clamp((this.scaling * (attEnv - susEnv)) / (susEnv + 1e-6), -1.5, 1.5);
|
||||
const attScale = peakiness > 0 ? peakiness : 0;
|
||||
const susScale = peakiness < 0 ? -peakiness : 0;
|
||||
const attackGain = dbToLin(this.attackAmt * attScale * 18);
|
||||
const sustainGain = dbToLin(this.sustainAmt * susScale * 36);
|
||||
const gain = clamp(attackGain * sustainGain, 0, 8);
|
||||
avgGain = lerp(avgGain, gain, this.gainCoeff);
|
||||
const makeup = avgGain > 1e-3 ? 1 / avgGain : 1;
|
||||
const wet = sample * gain * makeup;
|
||||
let y = lerp(sample, wet, this.mix);
|
||||
y /= 1 + Math.abs(y); // soft clip
|
||||
output[ch][n] = y;
|
||||
}
|
||||
this.attackEnv[ch] = attEnv;
|
||||
this.sustainEnv[ch] = susEnv;
|
||||
}
|
||||
this.avgGain = avgGain;
|
||||
return true;
|
||||
}
|
||||
}
|
||||
|
||||
registerProcessor('transient-processor', TransientProcessor);
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue