Introduction

Welcome to the I/O Developer Guide

Here we review the architectural principles and design philosophy.

I/O is built on a modular and extensible audio-processing graph (DAG), where each node fulfills a specific role and well-defined role —whether synthesis, effect/processing, analysis, or rendering— and communicates through a stable, deterministic and consistent data flow.

Throughout this guide, we explore how to create, connect, and configure nodes, including practical examples, videos, and audio captures.

Documentation (API)

Here you will find the complete documentation describing the public API in detail, along with use cases and recommended implementation patterns.

Access the community section to explore additional resources.

Introduction

This guide is designed to provide a complete understanding of the I/O ecosystem, from conceptual foundations to advanced practice. As you progress, you will find explanations, code examples, and audiovisual resources that will help you master the engine’s architecture.

If you plan to contribute to the project, we also recommend reviewing the sections dedicated to best practices, code style, and collaboration guidelines—essential for maintaining quality and consistency in team development promoting cohesive workflows

What is I/O?

I/O is a cross-platform audio engine specifically designed for advanced real-time processing through a directed acyclic graph (DAG) of highly configurable nodes.

Each node represents a functional audio unit—whether a generator, processor, analyzer, or output destination—and all interact within a shared graph context that manages synchronization, latency, and data flow with sample-accurate precision.

The system combines deterministic design principles, thread-safe execution, and a no dynamic allocation policy during render, guaranteeing stability under heavy conditions.

I/O enables building effect chains and complex spatialization environments while maintaining a consistent and deterministic render cycle, abstracting hardware drivers and OS layers through a unified API for realtime/offline processing.

I/O is implemented in Swift for greater expressiveness, safety and interoperability.

In summary, this is not merely a framework—it is an infrastructure for composing sound as if it were code, where every connection, node, and parameter forms part of a controlled, extensible, and synchronized network enabling precision over complex audio behaviors.