AURORA Audio Engine

Ultra-low-latency virtual audio routing, mixing and performance platform for Windows

---

About

AURORA Audio Engine is a real-time virtual audio routing and mixing platform built for Windows. It captures WASAPI loopback audio (browser, games, Spotify, DAWs) from up to two independent music sources, processes them through a 64-bit float DSP chain, and routes the result to headphones, monitors, and a virtual microphone — all with lock-free, real-time-safe audio threads.

It started as a tool for a single use case: in-game concert DJing. It is now a general platform for streamers, musicians, and power users.

Features

• Dual-track routing matrix — mix two music sources (Track A, Track B) with independent gain and 3-band EQ, crossfader, and a master chain.
• Real-time DSP in 64-bit float — 3-band EQ, compressor, noise gate, limiter, delay, reverb, ducking, auto-gain, and voice activation.
• Virtual microphone bridge — route processed audio to any app as a mic.
• WASAPI shared & exclusive modes — buffer sizes down to 64 frames (≈1.3 ms at 48 kHz) for pro-level low latency.
• Intelligent profiles — Rust Outpost, Discord, Streaming, Recording, Music Performance, and Balanced presets that configure the entire chain in one click.
• First-run hardware calibration — measures real round-trip latency and suggests the best buffer for your device.
• Real-time analysis — live spectrum, BPM detection, and latency readout.
• Clean-room engineered — windows FFI, lock-free SPSC ring buffers, and a fully test-covered DSP core.

Repository layout

aurora-audio-engine/
├── aurora-core/           # Core engine library (no GUI dependency)
│   ├── src/dsp/           #   EQ, dynamics, FX, analysis, voice
│   ├── src/wasapi/        #   WASAPI capture, render, mic, device enum
│   ├── src/engine.rs      #   Thread orchestration & DSP loop
│   ├── src/ring.rs        #   Lock-free SPSC ring buffer
│   └── src/{profile,calibrate,ai}.rs
├── aurora-gui/            # eframe/egui desktop app (bin: `aurora`)
├── docs/                  # Architecture, usage, developer docs
├── assets/                # Branding (logo, banner, icon)
└── .github/               # CI workflows & issue/PR templates

Getting started

Prerequisites

• Windows 10/11 (x64)
• Rust stable (1.87+, see rust-toolchain.toml)
• A physical microphone and audio device with WASAPI drivers

Build & run

# Build the entire workspace
cargo build --release

# Run the desktop app
cargo run --release --bin aurora

# Run the full test suite
cargo test --workspace

# Lint
cargo clippy --workspace --all-targets -- -D warnings

First run

1. Plug in your mic and headphones.
2. Launch aurora.
3. Run Calibration in the ENGINE/LATENCY panel — it measures your device's real round-trip latency and recommends a buffer.
4. Pick a profile (e.g. Rust Outpost for in-game music).
5. Route your playback device into Track A and select the virtual mic as the output target in the apps you want to feed.

Documentation

Document	Contents
Architecture	Module design, DSP chain, threading model
Usage guide	Profiles, routing, calibration, latency tuning
Developer guide	Building, testing, extending the engine
Contributing	How to contribute code
Security	Vulnerability reporting & policy
Roadmap	Planned work
Changelog	Release history

Status

Pre-release (v0.1.0). The DSP core is fully test-covered; WASAPI device interaction is validated against real Windows 10 hardware.

• Core engine, DSP chain, ring buffer
• Profiles, calibration, GUI
• Unit + integration test suite (67+ tests)
• CI pipeline (fmt, clippy, tests, release build, dependency audit)
• Virtual audio driver packaging (optional, for global system routing)
• Preset manager & shareable profile files
• ONNX-based AI voice intelligence

Roadmap

See docs/ROADMAP.md for the detailed plan, including a future system-wide virtual audio driver.

License

MIT — see LICENSE.

---

_{Built with Rust, `windows` crate, and `egui`/`eframe`.}