AudioMix 🎧

AI-powered DJ mixing engine. Upload your tracks, let the engine analyze BPM, key, energy, and semantic structure, then generate a studio-quality MP3 mix — with harmonic forcing, bass swaps, filter sweeps, and beat-synced crossfades.

---

Features

• Semantic section detection — classifies Intro / Verse / Pre-Chorus / Drop / Outro
• Pre-computed STFT pattern matching — finds the best entry point in O(1) per window
• Harmonic forcing — pitch-shifts tracks to a compatible Camelot key without changing tempo
• Bass swaps & HP filter sweeps — studio-grade DSP transitions
• Tension riser — loop-halving engine builds a chaotic riser into every drop
• Stem-aware crossfade — vocals exit early, instrumentals carry the transition
• In-memory export — no intermediate WAV files written to disk

---

Quickstart

Prerequisites

• Python 3.11+
• Node.js 20+
• ffmpeg in PATH (required by pydub)

1 — Clone

git clone https://github.com/sahiee-dev/audiomix.git
cd audiomix
cp .env.example .env

2 — Backend

cd backend
pip install -r app/requirements.txt
uvicorn app.main:app --reload --port 8000

3 — Frontend

cd frontend
npm install
npm run dev

Open http://localhost:5173 in your browser.

---

Docker (optional)

docker-compose up

Both services start automatically. Frontend at http://localhost:5173, API at http://localhost:8000.

---

API Reference

Method	Endpoint	Description
GET	/api/health	Liveness probe
POST	/api/upload	Upload audio files
POST	/api/analyze?session=<id>	Extract features
POST	/api/order	Sort tracks by mode
POST	/api/mix?session=<id>	Generate the mix
GET	/api/sessions	List active sessions
DELETE	/api/sessions/<id>	Delete a session
GET	/api/preview/<session>/<file>	Stream audio
GET	/api/download/<session>/<file>	Download file

Full interactive docs at http://localhost:8000/docs.

---

Developer Commands

make install        # install backend + frontend deps
make dev-backend    # start FastAPI with hot-reload
make dev-frontend   # start Vite dev server
make test           # run backend unit tests
make lint           # run ruff linter
make clean          # remove uploads and Python caches

---

Project Structure

audiomix/
├── backend/
│   └── app/
│       ├── main.py           # FastAPI app entry point
│       ├── api.py            # API router
│       ├── config.py         # Environment-based configuration
│       ├── schemas.py        # Pydantic request/response models
│       ├── mixer.py          # Stem-level mashup engine
│       ├── segmenter.py      # Semantic section classifier
│       ├── pattern_matcher.py# Pre-computed STFT matching
│       ├── pitch_engine.py   # Harmonic forcing via pyrubberband
│       ├── tension_looper.py # Beat-grid loop-halving riser
│       ├── dsp_fx.py         # EQ kills, bass swap, HP sweep
│       └── audio_utils.py    # Feature extraction utilities
└── frontend/
    └── src/
        ├── App.jsx
        ├── api.js            # API client
        └── components/

---

License

MIT