audio-slot

audio-slot

Web Audio API triggerable audio slot described with JSON consisting of sources, processors, and modulators.

Pass in an AudioContext extended with sources, processors, and modulators and get back a single AudioNode. Because the entire setup of AudioNodes is done with a JSON object, it makes it simple and easy to persist and synchronize audio setups across a network. Used to manage each slot in soundbank.

Install

$ npm install audio-slot

API

var AudioSlot = require('audio-slot')

AudioSlot(audioContext, descriptor)

audioContext: instance of AudioContext extended with sources, processors, and modulators lookups.

descriptor: an object describing the desired configuration of AudioNodes. See example below.

Returns a slot AudioNode.

slot.update(descriptor)

A new descriptor object to update the internal nodes to. Will find and update all changed params, add new nodes and remove any unused nodes.

slot.triggerOn(at, velocity)

Trigger all source and modulator start methods.

slot.triggerOff(at)

Recommend triggering stop() on sources at given time. This event may be ignored by some sources which specify their own off time such as oneshot samples.

slot.choke(at)

Cut off playback of slot sources and modulators immediately.

slot.connect(destination, inputNumber)

Connect the slot output to another AudioNode (optionally specifying inputNumber for nodes with multiple inputs)

slot.disconnect()

Disconnect the slot from all previously connected destination AudioNodes.

Example

var AudioSlot = require('audio-slot')

var audioContext = new webkitAudioContext()

audioContext.sources = {
  sample: require('soundbank-sample'),
  oscillator: require('soundbank-oscillator')
}

audioContext.processors = {
  gain: audioContext.createGain.bind(audioContext), // can use built in nodes
  filter: audioContext.createBiquadFilter.bind(audioContext),
  delay: require('soundbank-delay')
}

audioContext.modulators = {
  lfo: require('lfo'),
  adsr: require('adsr')
}

audioContext.sampleCache = {}

loadAudioBuffer('/sounds/hiss.wav', function(err, buffer){
  audioContext.sampleCache['hiss.wav'] = buffer
})

var slot = AudioSlot(audioContext, {
  sources: [
    { node: 'sample', // coresponds to key in audioContext.sources
      url: 'hiss.wav',
      startOffset: 0.2,
      endOffset: 1,
      transpose: -3,
      mode: 'loop',

      amp: { // add a modulator from audioContext.modulators
        node: 'adsr',
        value: 1,
        attack: 0.01,
        decay: 0.3,
        sustain: 0.6,
        release: 0.4
      }
    }
  ],
  processors: [
    { node: 'delay',
      delayTime: 100,
      wetLevel: 0.6
    }
  ]
})

slot.connect(audioContext.destination)

// schedule a 2s playback of looped sample
slot.triggerOn(audioContext.currentTime)
slot.triggerOff(audioContext.currentTime + 2)

You can also use audio-slot instances as routing busses as by default they accept inputs from other nodes.

var bus = AudioSlot(audioContext, {

  // specify how to handle trigger events
  inputMode: 'on' // other options: bypass, holdOn, holdOff, bypassOn, bypassOff

  processors: [
    { node: 'filter',
      type: 'lowpass',
      frequency: { // modulate the frequency AudioParam with an lfo
        node: 'lfo',
        rate: 2,
        amp: 100,
        value: 200
      }
    }
  ]
})

// route in another audio node

slot.connect(bus)
bus.connect(audioContext.destination)