Process Audio Capture

一个用于捕获和可视化指定 Windows 进程音频输出的应用。

简介

本应用通过一个 C++ 子进程实时捕获指定 Windows 进程的音频输出，并使用 Electron 和 React 构建的用户界面进行实时频谱可视化。进程间通信采用 stdio 标准流，以确保高效的数据传输和较低的性能损耗。

本项目旨替代以下项目：

• Capture-Audio-from-Process---javascript-addon
• win-process-audio-capture

功能特性

• 进程选择：自动列出当前正在播放音频的 Windows 进程。
• 实时音频捕获：从选定的进程中捕获原始 PCM 音频流。
• 频谱可视化：使用 Web Audio API 和 Canvas 实时显示音频频谱图。
• 高性能：通过优化的 C++ 后端和高效的 stdio 通信，最大限度地减少延迟和性能开销。

技术栈

• 前端: React, TypeScript, Vite, shadcn/ui, Tailwind CSS
• 应用外壳: Electron
• 音频捕获 (C++ 子进程):
  • C++17
  • Windows Core Audio APIs (WASAPI)
  • nlohmann/json 用于 IPC 消息序列化
  • Microsoft WIL 用于简化 WinAPI 调用

如何工作

应用由两部分组成：

1. C++ 子进程 (process-audio-capture.exe): 一个命令行程序，使用 WASAPI 对目标进程进行环回音频捕获。它负责枚举音频会话，并将捕获到的原始 PCM 音频数据输出到其标准输出流。
2. Electron 应用:
   • 主进程: 负责启动和管理 C++ 子进程，并通过 stdin/stdout 与其通信。它将捕获的音频数据通过 IPC 转发给渲染进程。
   • 渲染进程: 一个 React 应用，提供用户界面，允许用户选择目标进程、控制捕获的开始/停止，并使用 Web Audio API 对接收到的音频数据进行处理和可视化。

开发与运行

1. 编译 C++ 子进程

• C++ 项目位于 process-audio-capture/ 目录。
• 需要安装 Visual Studio 及“使用 C++ 的桌面开发”工作负载。
• 依赖项 nlohmann.json 和 wil 通过 NuGet 管理。在 Visual Studio 中打开 process-audio-capture.sln 即可自动还原依赖包。
• 编译项目（Debug 或 Release 模式），确保生成的可执行文件位于 process-audio-capture/x64/Debug/process-audio-capture.exe 或相应路径。

2. 运行 Electron 应用

首先，请确保已安装 Node.js 和 Yarn。

# 安装依赖
yarn

# 启动开发服务器
yarn dev

---

Process Audio Capture (English)

An application for capturing and visualizing audio output from specific Windows processes.

Introduction

This application captures the audio output of a specified Windows process in real-time using a C++ subprocess and provides a user interface built with Electron and React for live spectrum visualization. Communication between processes is handled via stdio to ensure efficient data transfer and low performance overhead.

This project aims to be a alternative to:

• Capture-Audio-from-Process---javascript-addon
• win-process-audio-capture

Features

• Process Selection: Automatically lists Windows processes that are currently playing audio.
• Real-time Audio Capture: Captures raw PCM audio stream from the selected process.
• Spectrum Visualization: Displays a real-time audio spectrum using the Web Audio API and Canvas.
• High Performance: Minimizes latency and performance impact through an optimized C++ backend and efficient stdio communication.

Tech Stack

• Frontend: React, TypeScript, Vite, shadcn/ui, Tailwind CSS
• Application Shell: Electron
• Audio Capture (C++ Subprocess):
  • C++17
  • Windows Core Audio APIs (WASAPI)
  • nlohmann/json for IPC message serialization
  • Microsoft WIL for simplified WinAPI calls

How It Works

The application consists of two main components:

1. C++ Subprocess (process-audio-capture.exe): A command-line program that uses WASAPI for loopback audio capture of a target process. It is responsible for enumerating audio sessions and writing the captured raw PCM audio data to its standard output.
2. Electron Application:
   • Main Process: Spawns and manages the C++ subprocess, communicating with it via stdin/stdout. It forwards the captured audio data to the renderer process via IPC.
   • Renderer Process: A React application that provides the UI, allowing the user to select a target process, control the start/stop of the capture, and visualize the received audio data using the Web Audio API.

Development & Usage

1. Build the C++ Subprocess

• The C++ project is located in the process-audio-capture/ directory.
• Visual Studio with the "Desktop development with C++" workload is required.
• Dependencies nlohmann.json and wil are managed via NuGet. Open process-audio-capture.sln in Visual Studio to restore packages automatically.
• Build the project (in Debug or Release mode). Ensure the output executable is located at process-audio-capture/x64/Debug/process-audio-capture.exe or the corresponding path.

2. Run the Electron App

First, ensure you have Node.js and Yarn installed.

# Install dependencies
yarn

# Run in development mode
yarn dev