Skip to main content
The Moonshine Voice repository includes comprehensive examples demonstrating how to integrate and use the library across multiple platforms and languages.

​
Example Structure

Examples are organized by platform in the examples/ directory:
  • Android - Mobile transcription app
  • C++ - Cross-platform minimal example
  • iOS - Native iOS transcription app
  • macOS - Desktop applications
  • Python - Various transcription and intent recognition examples
  • Raspberry Pi - IoT and robotics examples
  • Windows - Command-line transcriber

​
Python Examples

Python examples demonstrate the core features of Moonshine Voice and are located in examples/python/.

​
Basic Transcription

File: examples/python/basic_transcription.py This example demonstrates both streaming and non-streaming transcription:
  • Non-streaming mode: Transcribe a complete audio file offline
  • Streaming mode: Process audio chunks in real-time with live updates
  • Implements custom TranscriptEventListener to handle events
  • Shows how to use load_wav_file() to load audio data
python examples/python/basic_transcription.py --language en

​
Microphone Transcription

File: examples/python/mic_transcription.py Real-time transcription from your microphone:
  • Uses MicTranscriber for automatic microphone integration
  • Implements terminal-aware output (interactive vs. non-interactive)
  • Shows real-time text updates as you speak
  • Handles line completion and text updates
python examples/python/mic_transcription.py --language en

​
Intent Recognition

File: examples/python/intent_recognition.py Voice command recognition using semantic matching:
  • Demonstrates IntentRecognizer for natural language commands
  • Uses embedding models for fuzzy phrase matching
  • Registers multiple intents with custom handlers
  • Shows confidence scores for matches
Example commands:
  • “Turn on the lights” / “Switch on the lights”
  • “What’s the weather like?” / “Weather forecast”
  • “Set a timer” / “Start a timer for 5 minutes”
python examples/python/intent_recognition.py

​
Platform-Specific Examples

​
Android

Location: examples/android/Transcriber/ Full Android application demonstrating:
  • Integration with Android Studio and Gradle
  • Maven package integration
  • Real-time microphone transcription
  • Native Android UI with RecyclerView for transcript display
Download: android-examples.tar.gz

​
iOS

Location: examples/ios/Transcriber/ Native iOS app featuring:
  • Swift integration using Swift Package Manager
  • SwiftUI interface
  • Real-time microphone transcription
  • Model file bundling in app resources
Download: ios-examples.tar.gz

​
macOS

Two examples demonstrating different approaches:

​
Basic Transcription

Location: examples/macos/BasicTranscription/
  • Command-line tool for transcribing audio files
  • Shows Swift Package Manager integration

​
Mic Transcription

Location: examples/macos/MicTranscription/
  • Real-time microphone transcription for macOS
  • Native audio input handling
Download: macos-examples.tar.gz

​
Windows

Location: examples/windows/cli-transcriber/ Command-line transcriber for Windows:
  • Visual Studio project (.vcxproj)
  • C++ implementation using the high-level API
  • Includes download-lib.bat script to fetch pre-built libraries
Download: windows-examples.tar.gz 
cd examples\windows\cli-transcriber
.\download-lib.bat
msbuild cli-transcriber.sln /p:Configuration=Release /p:Platform=x64
python -m moonshine_voice.download --language en
x64\Release\cli-transcriber.exe --model-path <path> --model-arch <arch>

​
C++ (Cross-Platform)

Location: examples/c++/ Minimal, platform-independent example:
  • Uses the C++ header-only API (moonshine-cpp.h)
  • Shows basic transcription workflow
  • Demonstrates linking with pre-built libraries
  • Works on Linux, macOS, and Windows
Key Features:
  • No platform-specific dependencies
  • Simple build with g++ or clang++
  • Links against static library (.a or .lib)

​
Raspberry Pi

Location: examples/raspberry-pi/my-dalek/ File: my-dalek.py Robotics voice control example:
  • Voice commands for robot movement (forward, backward, left, right)
  • Intent recognition with natural language variations
  • Designed for Raspberry Pi with USB microphone
  • Demonstrates IoT device integration
Commands:
  • “Move forward” / “Go ahead” / “Advance”
  • “Turn left” / “Go left”
  • “Exterminate” / “Kill all humans” (for fun!)
pip install moonshine-voice
python examples/raspberry-pi/my-dalek/my-dalek.py
Download: raspberry-pi-examples.tar.gz

​
Common Patterns

All examples demonstrate these key patterns:
  1. Model Loading: Download and load appropriate models for your language
  2. Event Listeners: Implement TranscriptEventListener for handling transcription updates
  3. Audio Input: Feed audio data from files, microphones, or other sources
  4. Session Management: Start/stop sessions to control transcription lifecycle

​
Getting Started

  1. Choose the platform/language that matches your project
  2. Download the example archive or clone the repository
  3. Follow the platform-specific setup instructions
  4. Run the example to verify it works
  5. Adapt the code for your application

​
Next Steps

Build docs developers (and LLMs) love

Get started for freeTalk to us