WorkletProcessingNode

The WorkletProcessingNode interface represents a node in the audio processing graph that can process audio using a worklet function. Unlike WorkletNode which only provides read-only access to audio data, WorkletProcessingNode allows you to modify the audio signal by providing both input and output buffers.

This node lets you execute a worklet that receives input audio data and produces output audio data, making it perfect for creating custom audio effects, filters, and processors. The worklet processes the exact number of frames provided by the audio system in each call.

For more information about worklets, see our Introduction to worklets.

Constructor

BaseAudioContext.createWorkletProcessingNode(worklet, workletRuntime)

Example

import { AudioContext, AudioRecorder } from 'react-native-audio-api';

// This example shows how to create a simple gain effect using WorkletProcessingNode
function App() {
    const recorder = new AudioRecorder({
      sampleRate: 16000,
      bufferLengthInSamples: 16000,
    });

    const audioContext = new AudioContext({ sampleRate: 16000 });

    // Create a simple gain worklet that multiplies the input by a gain value
    const gainWorklet = (
        inputData: Array<Float32Array>,
        outputData: Array<Float32Array>,
        framesToProcess: number,
        currentTime: number
    ) => {
        'worklet';
        const gain = 0.5; // 50% volume

        for (let ch = 0; ch < inputData.length; ch++) {
            const input = inputData[ch];
            const output = outputData[ch];

            for (let i = 0; i < framesToProcess; i++) {
                output[i] = input[i] * gain;
            }
        }
    };

    const workletProcessingNode = audioContext.createWorkletProcessingNode(
        gainWorklet,
        'AudioRuntime'
    );
    const adapterNode = audioContext.createRecorderAdapter();

    adapterNode.connect(workletProcessingNode);
    workletProcessingNode.connect(audioContext.destination);
    recorder.connect(adapterNode);
    recorder.start();
}
}

Worklet Parameters Explanation

The worklet function receives four parameters:

inputData: Array<Float32Array>

A two-dimensional array where:

• First dimension represents the audio channel (0 = left, 1 = right for stereo)
• Second dimension contains the input audio samples for that channel
• You should read from these buffers to get the input audio data
• The length of each Float32Array equals the framesToProcess parameter

outputData: Array<Float32Array>

A two-dimensional array where:

• First dimension represents the audio channel (0 = left, 1 = right for stereo)
• Second dimension contains the output audio samples for that channel
• You must write to these buffers to produce the processed audio output
• The length of each Float32Array equals the framesToProcess parameter

framesToProcess: number

The number of audio samples to process in this call. This determines how many samples you need to process in each channel's buffer. This value will be at most 128.

currentTime: number

The current audio context time in seconds when this worklet call begins. This represents the absolute time since the audio context was created.

Audio Processing Pattern

A typical WorkletProcessingNode worklet follows this pattern:

const audioProcessor = (
    inputData: Array<Float32Array>,
    outputData: Array<Float32Array>,
    framesToProcess: number,
    currentTime: number
) => {
    'worklet';

    for (let channel = 0; channel < inputData.length; channel++) {
        const input = inputData[channel];
        const output = outputData[channel];

        for (let sample = 0; sample < framesToProcess; sample++) {
            // Process each sample
            // Read from: input[sample]
            // Write to: output[sample]
            output[sample] = processAudioSample(input[sample]);
        }
    }
};

Properties

It has no own properties but inherits from AudioNode.

Methods

It has no own methods but inherits from AudioNode.

Performance Considerations

Since WorkletProcessingNode processes audio in real-time, performance is critical:

• Keep worklet functions lightweight and efficient
• Avoid complex calculations that could cause audio dropouts
• Process samples in-place when possible
• Consider using lookup tables for expensive operations
• Use AudioRuntime for better performance, UIRuntime for UI integration
• Test on target devices to ensure smooth audio processing

Use Cases

• Audio Effects: Reverb, delay, distortion, filters
• Audio Processing: Compression, limiting, normalization
• Real-time Filters: EQ, high-pass, low-pass, band-pass filters
• Custom Algorithms: Noise reduction, pitch shifting, spectral processing
• Signal Analysis: Feature extraction while passing audio through