moonshine.cpp

Standalone C++ implementation of Moonshine ASR with ONNXRuntime and no other dependencies.

Table of Contents

• Introduction
• Build Instructions
• Example Usage
• Using as a Library
• Credits
• License

Introduction

Moonshine ASR is an Automatic Speech Recognition (ASR) system implemented in C++ using ONNX Runtime. This project provides a standalone implementation that can be built and run on various platforms.

Build Instructions

Prerequisites

• CMake 3.15 or higher
• A C++17 compatible compiler
• ONNX Runtime for the building OS will be fetched in build time.

Building on Windows

1. Open a PowerShell terminal.

2. Navigate to the root directory of the project.

3. Run the build script:

   .\scripts\build.ps1 -BuildType Release

Building on Linux/MacOS

1. Open a terminal.

2. Navigate to the root directory of the project.

3. Run the build script:

   ./scripts/build.sh

Manual Build Steps

1. Create a build directory:

   mkdir -p build
   cd build

2. Configure the project with CMake:

   cmake -DCMAKE_BUILD_TYPE=Release ..

3. Build the project:

   cmake --build .

4. Install the project:

   cmake --install . --prefix ../dist

Build results

After building you should have a folder like so (e.g. on Windows):

./dist
├───bin
│       moonshine_example.exe
│       moonshine_live.exe
│       onnxruntime.dll
│       onnxruntime_providers_shared.dll
│       
├───include
│   └───moonshine
│           moonshine.hpp
│
└───lib
    │   moonshine.lib
    │   onnxruntime.dll
    │   onnxruntime_providers_shared.dll
    │
    └───cmake
        └───moonshine
                moonshine-config-version.cmake
                moonshine-config.cmake
                moonshine-targets-release.cmake
                moonshine-targets.cmake

Which will allow you to link against moonshine.cpp with a CMake find_package().

Example Usage

The project includes two example applications:

• moonshine_example: File-based transcription
• moonshine_live: Real-time microphone transcription (requires SDL2)

Building Examples

Build the project with examples enabled:

.\scripts\build.ps1 -BuildType Release -BuildExamples

File-based Transcription

Transcribe audio from a WAV file:

.\dist\bin\moonshine_example.exe <models_dir> <wav_file>

Replace <models_dir> with the directory containing your ONNX models and <wav_file> with the path to a WAV file.

Example:

.\dist\bin\moonshine_example.exe models\ example.wav

Live Transcription

Real-time transcription from microphone input:

.\dist\bin\moonshine_live.exe <models_dir>

Example:

.\dist\bin\moonshine_live.exe models\

Press ESC or q to stop recording and exit.

Using as a Library

To use Moonshine ASR as a library in your C++ project, follow these steps:

1. Build and install the Moonshine ASR library as described in the Build Instructions.

2. Link against the installed library in your CMake project:

   cmake_minimum_required(VERSION 3.15)
   project(my_project)
   
   # Find the Moonshine package
   find_package(Moonshine REQUIRED)
   
   add_executable(my_project main.cpp)
   
   # Link against the Moonshine library
   target_link_libraries(my_project PRIVATE Moonshine::moonshine)

3. Include the Moonshine header in your source code:

   #include <moonshine.hpp>
   
   int main() {
       MoonshineModel model("path/to/models");
       // Use the model...
       return 0;
   }

Credits

This project is based on the Moonshine ASR system: https://github.com/usefulsensors/moonshine. Special thanks to the Moonshine team for their contributions.

License

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