@affectively/wasm-audio-processor

High-performance WebAssembly audio processing utilities written in Rust.

Features

• Mix Audio Streams - Blend two audio streams with configurable volumes and fade effects
• Volume Control - Efficiently reduce audio volume
• Silence Generation - Create silence buffers of specified duration
• mu-law Codec - Encode/decode mu-law audio format
• Base64 I/O - Accept and return base64-encoded audio for easy integration

Installation

npm (WebAssembly)

npm install @affectively/wasm-audio-processor
# or
yarn add @affectively/wasm-audio-processor
# or 
bun add @affectively/wasm-audio-processor

Cargo (Rust)

[dependencies]
affectively-audio-processor = "1.0"

Quick Start

JavaScript/TypeScript

import init, { 
 mix_audio_streams, 
 reduce_volume, 
 create_silence,
 AudioMixerConfig 
} from '@affectively/wasm-audio-processor';

// Initialize WASM module
await init();

// Mix two audio streams
const config = new AudioMixerConfig(
 0.3, // whisper volume (0-1)
 1.0, // original volume (0-1)
 100, // fade in (ms)
 100, // fade out (ms)
 8000 // sample rate
);

const mixedAudio = mix_audio_streams(originalBase64, whisperBase64, config);

// Reduce volume
const quieterAudio = reduce_volume(audioBase64, 0.5);

// Create 500ms of silence
const silence = create_silence(500, 8000);

Rust

use affectively_audio_processor::{mix_audio_streams, reduce_volume, create_silence};

// Mix audio streams
let mixed = mix_audio_streams(original, whisper, &config);

// Reduce volume by 50%
let quiet = reduce_volume(audio, 0.5);

// Create 500ms silence at 8kHz
let silence = create_silence(500.0, 8000.0);

API Reference

mix_audio_streams(original, whisper, config)

Mix two base64-encoded mu-law audio streams.

Parameters:

• original: string - Base64-encoded original audio
• whisper: string - Base64-encoded audio to overlay
• config: AudioMixerConfig - Mixing configuration

Returns: string - Base64-encoded mixed audio

AudioMixerConfig

Configuration for audio mixing:

new AudioMixerConfig(
 whisperVolume: number, // Volume for whisper stream (0-1)
 originalVolume: number, // Volume for original stream (0-1)
 fadeInMs: number, // Fade-in duration in milliseconds
 fadeOutMs: number, // Fade-out duration in milliseconds
 sampleRate: number // Audio sample rate (e.g., 8000, 16000)
)

reduce_volume(audio, volume)

Reduce the volume of audio.

Parameters:

• audio: string - Base64-encoded mu-law audio
• volume: number - Volume multiplier (0-1)

Returns: string - Base64-encoded audio with reduced volume

create_silence(durationMs, sampleRate)

Create a silence buffer.

Parameters:

• durationMs: number - Duration in milliseconds
• sampleRate: number - Sample rate

Returns: string - Base64-encoded silence

Performance

Benchmarks compared to pure JavaScript implementations:

| Operation | JavaScript | WASM | Speedup | |-----------|------------|------|---------| | Mix 1s audio | 45ms | 18ms | 2.5x | | Volume reduction | 12ms | 4ms | 3x | | Create silence | 8ms | 2ms | 4x |

Building from Source

Prerequisites

• Rust 1.70+
• wasm-pack

Build

# For web
npm run build

# For Node.js
npm run build:node

# For bundlers (webpack, vite, etc.)
npm run build:bundler

License

MIT License - see LICENSE for details.

Related Packages

• @affectively/behavioral-taxonomy - Emotion & behavior datasets
• @affectively/utils - Utility functions

Made with ️ by AFFECTIVELY