docs/examples.md
Try the live demos in your browser at moonshine.ai.
The examples folder has code samples organized by platform. We use the usual tooling per stack (Android Studio and Gradle, Xcode and Swift on Apple platforms, Visual Studio on Windows). GitHub Releases currently ship the downloadable assets below (example trees are mostly named {platform}-{Project}.tar.gz; Windows and C++ also include prebuilt native library bundles).
To add the library to your own app rather than starting from a sample, see Adding the Library.
=== "Javascript"
Self-contained archives: extract, run `node serve.mjs`, then open the demo path in a browser.
- **[stt](https://github.com/moonshine-ai/moonshine/releases/latest/download/web-stt.tar.gz)** — Real-time speech recognition in the browser via WebAssembly; nothing is uploaded.
- **[tts](https://github.com/moonshine-ai/moonshine/releases/latest/download/web-tts.tar.gz)** — On-device speech synthesis and zero-shot voice cloning in the browser.
- **[agent-flow](https://github.com/moonshine-ai/moonshine/releases/latest/download/web-agent-flow.tar.gz)** — A multi-step spoken conversation (wifi setup) triggered by meaning rather than exact wording.
- **[dictation](https://github.com/moonshine-ai/moonshine/releases/latest/download/web-dictation.tar.gz)** — Dictate a document in the browser with on-device recognition.
- **[meeting-notes](https://github.com/moonshine-ai/moonshine/releases/latest/download/web-meeting-notes.tar.gz)** — Transcribe a meeting and your microphone separately, entirely on-device.
=== "Python"
Scripts live under [`examples/python/`](https://github.com/moonshine-ai/moonshine/blob/main/examples/python/). Install with `pip install moonshine-voice` first.
- **[basic_transcription.py](https://github.com/moonshine-ai/moonshine/blob/main/examples/python/basic_transcription.py)** — Transcribe a WAV file offline or with streaming events.
- **[mic_transcription.py](https://github.com/moonshine-ai/moonshine/blob/main/examples/python/mic_transcription.py)** — Live microphone transcription with in-place partial updates in the terminal.
- **[text_to_speech.py](https://github.com/moonshine-ai/moonshine/blob/main/examples/python/text_to_speech.py)** — Speak text aloud, optionally in a cloned voice.
- **[agent_flow.py](https://github.com/moonshine-ai/moonshine/blob/main/examples/python/agent_flow.py)** — Generator-based conversational agent that walks through wifi setup by voice or keyboard.
- **[ollama_voice.py](https://github.com/moonshine-ai/moonshine/blob/main/examples/python/ollama-voice/ollama_voice.py)** — Pipe finalized mic transcripts into a local [Ollama](https://ollama.com/) LLM and stream the reply.
=== "iOS"
Open the extracted Xcode project after downloading. Samples pull **`MoonshineVoice`** from the Swift package.
- **[Transcriber](https://github.com/moonshine-ai/moonshine/releases/latest/download/ios-Transcriber.tar.gz)** — Live microphone transcription UI with partial and final results.
- **[TextToSpeech](https://github.com/moonshine-ai/moonshine/releases/latest/download/ios-TextToSpeech.tar.gz)** — Speak text with on-device TTS and optional voice cloning.
- **[AgentFlow](https://github.com/moonshine-ai/moonshine/releases/latest/download/ios-AgentFlow.tar.gz)** — Spoken multi-step wifi-setup conversation using `AgentFlow`.
=== "Android"
Open the extracted folder in Android Studio. Samples depend on **`ai.moonshine:moonshine-voice:0.1.2`** from Maven Central.
- **[Transcriber](https://github.com/moonshine-ai/moonshine/releases/latest/download/android-Transcriber.tar.gz)** — Live microphone transcription UI with partial and final results.
- **[TextToSpeech](https://github.com/moonshine-ai/moonshine/releases/latest/download/android-TextToSpeech.tar.gz)** — Speak text with on-device TTS and optional voice cloning.
- **[AgentFlow](https://github.com/moonshine-ai/moonshine/releases/latest/download/android-AgentFlow.tar.gz)** — Spoken multi-step wifi-setup conversation using `AgentFlow`.
=== "Linux"
Portable C++ samples under [`examples/c++/`](https://github.com/moonshine-ai/moonshine/blob/main/examples/c++/README.md). The release archive includes sources and a `download-library.sh` helper for the prebuilt library, English model, and sample audio.
- **[cpp-examples.tar.gz](https://github.com/moonshine-ai/moonshine/releases/latest/download/cpp-examples.tar.gz)** — Full example tree plus the download helper for library and models.
- **[transcriber.cpp](https://github.com/moonshine-ai/moonshine/blob/main/examples/c++/transcriber.cpp)** — Minimal CLI that streams a WAV through the C++ API and prints the transcript.
- **[text-to-speech.cpp](https://github.com/moonshine-ai/moonshine/blob/main/examples/c++/text-to-speech.cpp)** — Minimal CLI that synthesizes speech from text and writes a WAV.
=== "MacOS"
Open the extracted Xcode project (or build with SwiftPM where noted). Samples pull **`MoonshineVoice`** from the Swift package.
- **[BasicTranscription](https://github.com/moonshine-ai/moonshine/releases/latest/download/macos-BasicTranscription.tar.gz)** — CLI that transcribes a WAV file, offline or with streaming events.
- **[MicTranscription](https://github.com/moonshine-ai/moonshine/releases/latest/download/macos-MicTranscription.tar.gz)** — Live microphone transcription from the command line.
- **[TextToSpeech](https://github.com/moonshine-ai/moonshine/releases/latest/download/macos-TextToSpeech.tar.gz)** — Speak text with on-device TTS and optional voice cloning.
- **[AgentFlow](https://github.com/moonshine-ai/moonshine/releases/latest/download/macos-AgentFlow.tar.gz)** — Spoken wifi-setup dialog flow (`AgentFlow`), with optional `--text` mode.
=== "Windows"
Self-contained Visual Studio project with the library, model, and sample WAV bundled.
- **[cli-transcriber](https://github.com/moonshine-ai/moonshine/releases/latest/download/windows-cli-transcriber.tar.gz)** — C++ CLI that transcribes a WAV or the live microphone; open in Visual Studio and build with F7.
=== "Raspberry Pi"
USB microphone recommended. Python samples use the same `moonshine-voice` pip package optimized for the Pi.
- **[my-dalek](https://github.com/moonshine-ai/moonshine/releases/latest/download/raspberry-pi-my-dalek.tar.gz)** — Playful voice-command interface for driving a robot-style script from the Pi.
- **[Pi Help Bot](https://github.com/moonshine-ai/pi-help-bot/archive/refs/heads/main.zip)** — A more advanced Raspberry Pi voice assistant ([source](https://github.com/moonshine-ai/pi-help-bot)).
None of the samples bundle model weights. Every engine downloads what it needs on first use — the speech model for Transcriber, the voice and G2P assets for TextToSpeech, all three plus the embedding model for AgentFlow — from https://download.moonshine.ai/, reporting progress through the onProgress callback the examples wire up to a label. Downloads are cached (under filesDir on Android, Caches/MoonshineModels on Apple platforms), so later launches run offline. Switching to a different voice triggers the same on-demand download for whatever that voice needs.
If you want a fully offline build with no first-run download, fetch the assets ahead of time and point the engine at them with modelsFrom(path); see docs/design/api-comparison.md for the tradeoff.