Skip to content

gajanansr/audioEngine

BranchesTags

Repository files navigation

🎙️ AutoMix

AI-Powered Audio Engine for Studio-Quality Vocals

Transform phone-recorded vocals into professional, studio-quality songs — no mixing or mastering knowledge required.

Features • What's Different • Quick Start • Architecture • Contributing

🚀 What is AutoMix?

AutoMix is a web-based AI audio engine designed specifically for beginners who want to create professional-sounding music without years of audio engineering experience. Upload your phone-recorded vocals, and our AI handles everything — from noise reduction to pitch correction to professional mastering.

✨ What Makes AutoMix Different?

Vs. Traditional DAWs (Logic, Ableton, FL Studio)

Feature Traditional DAWs AutoMix
Learning Curve Months to years Minutes
Plugin Knowledge 50+ plugins to master Zero plugins to learn
Mixing Expertise Required AI handles it
Price $200-$600+ Free/Affordable
Platform Desktop only Web-based, any device

Vs. Auto-Tune Apps (Voloco, StarMaker)

Feature Auto-Tune Apps AutoMix
Processing Pitch only Full vocal chain
Quality Basic Studio-grade
Intelligence Rule-based AI-adaptive
Output Processed vocals Complete mixed song

🎯 Key Differentiators

  1. 🧠 AI-Powered Smart Mixing
    Unlike simple auto-tune apps, AutoMix analyzes your vocal characteristics and automatically applies a complete professional mixing chain — EQ, compression, de-essing, reverb, and more — optimized for YOUR voice.

  2. 📱 Phone Recording Specialist
    Built specifically to rescue phone recordings. Our AI detects and fixes common phone recording issues: room noise, harsh frequencies, inconsistent levels, and more.

  3. 🎚️ Simplified Macro Controls
    Instead of 100+ confusing parameters, users get just 4 intuitive sliders:

    • AutoTune Strength — Natural to T-Pain
    • Reverb Amount — Dry to ambient
    • Polish Level — Raw to studio sheen
    • Vocal Loudness — Quiet to punchy

  4. 🔬 Reference Track Matching
    Upload a reference song you love, and AutoMix will analyze its vocal tone and match your mix to that professional sound.

  5. ⚡ Real-Time Preview + Cloud Render
    Hear changes instantly in your browser, then render final high-quality audio in the cloud with zero compression.

🔧 Features

Core Audio Processing

  • ✅ Intelligent Gain Staging — Auto-normalize to optimal levels
  • ✅ Noise Reduction — Neural network-powered (RNNoise)
  • ✅ Dynamic EQ — Subtractive (mud/harshness removal) and additive (presence/air)
  • ✅ Multi-stage Compression — Leveling + Control compression
  • ✅ Professional De-Esser — Dynamic sibilance control
  • ✅ AutoTune — Key-aware pitch correction with humanization
  • ✅ Saturation — Tube harmonics for warmth
  • ✅ Reverb & Delay — Space and dimension
  • ✅ Mastering Chain — Multiband compression + true-peak limiting

AI Features

  • 🎵 Automatic key and scale detection
  • 👤 Gender-aware processing (different EQ curves)
  • 🎤 Phone recording quality detection
  • 📊 Reference track tonal matching
  • 🎛️ Smart parameter optimization

🏗️ Architecture

┌────────────────────────────────────────────────────────┐
│                    CLIENT (PWA)                         │
│  React Frontend + Web Audio API + AudioWorklet         │
├────────────────────────────────────────────────────────┤
│                         ↓                               │
├────────────────────────────────────────────────────────┤
│                    BACKEND API                          │
│  Supabase (Auth + DB + Storage + Edge Functions)       │
├────────────────────────────────────────────────────────┤
│                         ↓                               │
├────────────────────────────────────────────────────────┤
│              WORKER (Cloud Render)                      │
│  Docker container for offline high-quality rendering   │
└────────────────────────────────────────────────────────┘

Tech Stack

Component Technology
Frontend React 18, TypeScript, Vite
Audio Engine Web Audio API, AudioWorklet
Backend Supabase (Postgres, Auth, Storage, Edge Functions)
Workers Docker, Node.js
Pitch Detection Essentia.js / Custom algorithms
Noise Reduction RNNoise (WASM)

🚀 Quick Start

Prerequisites

  • Node.js 18+
  • npm or yarn
  • Supabase account (free tier works)

Installation

# Clone the repository
git clone https://github.com/gajanansr/audioEngine.git
cd audioEngine

# Install dependencies
npm install

# Copy environment template
cp .env.example .env
# Fill in your Supabase credentials in .env

# Build the audio engine
npm run build

# Start the web app
cd web
npm install
npm run dev

The app will be available at http://localhost:5173

Environment Variables

# Supabase
SUPABASE_URL=https://your-project.supabase.co
SUPABASE_ANON_KEY=your-anon-key
SUPABASE_SERVICE_KEY=your-service-key

# Worker (optional, for cloud rendering)
POLL_INTERVAL_MS=5000
WORKER_ID=worker-1

📁 Project Structure

audioEngine/
├── src/                    # Core audio engine (TypeScript)
│   ├── core/               # Audio graph management
│   ├── plugins/            # DSP plugins (EQ, Compressor, etc.)
│   │   ├── dynamics/       # Compressor, Limiter, De-Esser
│   │   ├── eq/             # Parametric EQ
│   │   ├── effects/        # Reverb, Delay, Saturation
│   │   └── pitch/          # AutoTune, Pitch Detection
│   ├── ai/                 # AI analyzers & optimizers
│   ├── chains/             # Vocal chain configurations
│   └── macros/             # User-facing macro controls
├── web/                    # React frontend
│   └── src/
│       ├── components/     # UI components
│       ├── pages/          # Route pages
│       └── hooks/          # Custom React hooks
├── worker/                 # Cloud rendering worker
├── supabase/               # Database schema & types
└── docs/                   # Documentation

🤝 Contributing

We welcome contributions from developers of all skill levels! Here's how you can help:

Ways to Contribute

Type Description
🐛 Bug Reports Found a bug? Open an issue with steps to reproduce
💡 Feature Requests Have an idea? Open a discussion or issue
📝 Documentation Improve README, add examples, write tutorials
🔧 Code Pick an issue and submit a PR
🧪 Testing Write tests, find edge cases
🎨 Design UI/UX improvements

Development Setup

  1. Fork the repository

  2. Clone your fork

    git clone https://github.com/YOUR_USERNAME/audioEngine.git
cd audioEngine

  3. Create a branch

    git checkout -b feature/your-feature-name

  4. Install dependencies

    npm install
cd web && npm install

  5. Make your changes

    • Follow the existing code style
    • Add tests for new features
    • Update documentation as needed

  6. Run tests

    npm run test
npm run lint

  7. Commit your changes

    git add .
git commit -m "feat: add your feature description"

    We follow Conventional Commits:

    • feat: New feature
    • fix: Bug fix
    • docs: Documentation
    • refactor: Code refactoring
    • test: Adding tests
    • chore: Maintenance

  8. Push and create PR

    git push origin feature/your-feature-name

    Then open a Pull Request on GitHub.

Good First Issues

Look for issues labeled good first issue — these are great for newcomers:

  • 📚 Documentation improvements
  • 🧪 Adding unit tests
  • 🎨 UI polish
  • 🐛 Small bug fixes

Code Style

  • TypeScript for all code
  • ESLint for linting
  • Prettier for formatting (coming soon)
  • Meaningful variable/function names
  • Comments for complex DSP algorithms

Plugin Development

Want to add a new audio plugin? Here's the interface to implement:

interface AudioPlugin {
  id: string;
  name: string;
  type: 'insert' | 'send';
  bypass: boolean;
  
  process(input: Float32Array, output: Float32Array): void;
  setParameter(name: string, value: number): void;
  getParameter(name: string): number;
  getParameterDescriptors(): ParameterDescriptor[];
}

See src/plugins/base/BasePlugin.ts for the base class, and check existing plugins for examples.

🗺️ Roadmap

  • Core audio graph manager
  • Basic plugins (EQ, Compressor, Limiter)
  • AutoTune module
  • AI parameter optimizer
  • Macro control system
  • WASM pitch correction (higher quality)
  • RNNoise integration
  • Reference track matching
  • Cloud rendering pipeline
  • Mobile-optimized UI
  • Preset system

📄 License

This project is licensed under the MIT License — see the LICENSE file for details.

🙏 Acknowledgments

📬 Contact

Have questions? Open an issue or reach out!

Made with ❤️ for musicians who just want to sound good

About

AutoMix is an AI-powered web-based audio engine that helps beginners turn raw, phone-recorded vocals into professional, studio-quality songs - without needing mixing or mastering knowledge.

automix-audio.vercel.app

Resources

Readme
Activity

Stars

1 star

Watchers

0 watching

Forks

0 forks
Report repository

Releases

1 tags

Packages

 
 
 

Contributors

Languages