Add installer script (Inno Setup) and build instructions

Author: seyeong-han

.gitignore

@@ -68,4 +68,5 @@ project.nuget.cache
 *.nuget.g.props
 *.nuget.g.targets
 publish/
+installer-output/
 .vs/

voxtral_realtime/windows/README.md

@@ -4,104 +4,101 @@ Real-time speech transcription desktop app powered by ExecuTorch with CUDA accel
 
 This is the Windows equivalent of the [macOS Voxtral Realtime app](../macos/).
 
-## Quick Start (Pre-built Release)
+## Quick Start
 
-Download `VoxtralRealtime.exe` from the [Releases](https://github.com/meta-pytorch/executorch-examples/releases) page and run it directly. No installation required.
+Download `VoxtralRealtime-Setup.exe` from the [Releases](https://github.com/meta-pytorch/executorch-examples/releases) page and run the installer. Everything is bundled -- the app, runner, model weights, and tokenizer. No additional downloads required.
 
-You also need:
-- The `voxtral_realtime_runner.exe` (built from ExecuTorch with CUDA support)
-- Model files from HuggingFace (see [Model Files](#model-files) below)
+After install, launch from the Start Menu or desktop shortcut and click "Start Transcription".
 
-## Prerequisites
+### Requirements
 
 - Windows 10/11 with NVIDIA GPU (CUDA-capable)
 - CUDA Toolkit installed (auto-detected from `C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\`)
-- [.NET 8.0 SDK](https://dotnet.microsoft.com/download/dotnet/8.0) (only for building from source)
 
-## Model Files
+## Features
 
-Download from HuggingFace:
+- **Live Transcription** - Start/Pause/Resume with streaming text output
+- **Session Management** - Save, rename, pin, delete, and export sessions (TXT/JSON/SRT)
+- **Text Replacements** - Auto-correct transcription (e.g., "executorch" -> "ExecuTorch")
+- **Text Snippets** - Voice-triggered templates for common text blocks
+- **Dictation Mode** - Ctrl+Space global hotkey, floating overlay, auto-paste to any app, auto-stop on 2s silence
+- **Audio Level Visualization** - Real-time waveform display
 
-```powershell
-pip install huggingface_hub
-huggingface-cli download younghan-meta/Voxtral-Mini-4B-Realtime-2602-ExecuTorch-CUDA --local-dir voxtral_rt_exports
-```
+## Keyboard Shortcuts
 
-This downloads: `model.pte`, `preprocessor.pte`, `aoti_cuda_blob.ptd`
+| Shortcut | Action |
+|----------|--------|
+| Ctrl+Shift+R | Start / Resume transcription |
+| Ctrl+. | Pause transcription |
+| Ctrl+Enter | End session |
+| Ctrl+Space | Toggle dictation mode |
+
+## Build from Source
 
-You also need the tokenizer from the base model:
+For developers who want to build the app themselves.
+
+### Prerequisites
+
+- [.NET 8.0 SDK](https://dotnet.microsoft.com/download/dotnet/8.0)
+- Pre-built `voxtral_realtime_runner.exe` (see [Building the Runner](#building-the-runner))
+- Model files from HuggingFace (see [Model Files](#model-files))
+
+### Model Files
 
 ```powershell
+pip install huggingface_hub
+huggingface-cli download younghan-meta/Voxtral-Mini-4B-Realtime-2602-ExecuTorch-CUDA-Windows --local-dir voxtral_rt_exports
 huggingface-cli download mistralai/Voxtral-Mini-4B-Realtime-2602 tekken.json --local-dir voxtral_tokenizer
 ```
 
-## Building the Runner
-
-Build the `voxtral_realtime_runner.exe` from the ExecuTorch repo:
+### Building the Runner
 
 ```bash
 cd executorch
 cmake --preset voxtral-realtime-cuda
 cmake --build --preset voxtral-realtime-cuda
 ```
 
-The runner will be at `cmake-out/examples/models/voxtral_realtime/Release/voxtral_realtime_runner.exe`.
-
-## Build from Source
+### Build and Run
 
 ```powershell
-# Install .NET SDK if not already installed
-winget install Microsoft.DotNet.SDK.8
-
-# Build
 cd VoxtralRealtime
 dotnet restore
 dotnet build --configuration Release
-
-# Run
 dotnet run --project VoxtralRealtime --configuration Release
 ```
 
-## Publish Standalone Executable
-
-Create a single self-contained exe (no .NET runtime required on target machine):
+### Publish Standalone Executable
 
 ```powershell
 cd VoxtralRealtime
 dotnet publish VoxtralRealtime --configuration Release --runtime win-x64 --self-contained true /p:PublishSingleFile=true /p:IncludeNativeLibrariesForSelfExtract=true /p:DebugType=none -o publish
 ```
 
-The output `publish\VoxtralRealtime.exe` can be distributed and run on any Windows x64 machine.
-
-## Configuration
-
-On first launch, the app auto-loads the model from default paths. All paths are configurable in Settings:
+### Create Installer
 
-| File | Default Path |
-|------|-------------|
-| Runner | `cmake-out\examples\models\voxtral_realtime\Release\voxtral_realtime_runner.exe` |
-| Model | `voxtral_rt_exports_wsl\model.pte` |
-| Preprocessor | `voxtral_rt_exports_wsl\preprocessor.pte` |
-| CUDA blob | `voxtral_rt_exports_wsl\aoti_cuda_blob.ptd` |
-| Tokenizer | `tekken.json` |
+Builds a self-contained installer that bundles the app, runner, model weights, and tokenizer:
 
-## Features
+```powershell
+# 1. Install Inno Setup (one-time)
+winget install JRSoftware.InnoSetup
 
-- **Live Transcription** - Start/Pause/Resume with streaming text output
-- **Session Management** - Save, rename, pin, delete, and export sessions (TXT/JSON/SRT)
-- **Text Replacements** - Auto-correct transcription (e.g., "executorch" -> "ExecuTorch")
-- **Text Snippets** - Voice-triggered templates for common text blocks
-- **Dictation Mode** - Ctrl+Space global hotkey, floating overlay, auto-paste to any app, auto-stop on 2s silence
-- **Audio Level Visualization** - Real-time waveform display
+# 2. Publish the app
+cd VoxtralRealtime
+dotnet publish VoxtralRealtime --configuration Release --runtime win-x64 --self-contained true /p:PublishSingleFile=true /p:IncludeNativeLibrariesForSelfExtract=true /p:DebugType=none -o publish
 
-## Keyboard Shortcuts
+# 3. Build the installer
+cd ..
+ISCC installer.iss
+```
 
-| Shortcut | Action |
-|----------|--------|
-| Ctrl+Shift+R | Start / Resume transcription |
-| Ctrl+. | Pause transcription |
-| Ctrl+Enter | End session |
-| Ctrl+Space | Toggle dictation mode |
+The output `installer-output\VoxtralRealtime-Setup.exe` includes:
+- App executable (self-contained, no .NET runtime needed)
+- `voxtral_realtime_runner.exe` + `aoti_cuda_shims.dll`
+- Model weights (`model.pte`, `preprocessor.pte`, `aoti_cuda_blob.ptd`)
+- Tokenizer (`tekken.json`)
+- Start Menu and optional desktop shortcuts
+- Clean uninstall via Windows Settings
 
 ## Architecture
 

voxtral_realtime/windows/VoxtralRealtime/VoxtralRealtime/Resources/app.ico

@@ -8,6 +8,7 @@
     <AssemblyName>VoxtralRealtime</AssemblyName>
     <Nullable>enable</Nullable>
     <ImplicitUsings>enable</ImplicitUsings>
+    <ApplicationIcon>Resources\app.ico</ApplicationIcon>
   </PropertyGroup>
 
   <ItemGroup>

voxtral_realtime/windows/VoxtralRealtime/VoxtralRealtime/VoxtralRealtime.csproj

@@ -0,0 +1,33 @@
+; Voxtral Realtime Windows Installer - Inno Setup Script
+; Download Inno Setup from https://jrsoftware.org/isinfo.php
+
+[Setup]
+AppName=Voxtral Realtime
+AppVersion=1.0.0
+AppPublisher=Meta Platforms
+AppPublisherURL=https://github.com/meta-pytorch/executorch-examples
+DefaultDirName={autopf}\VoxtralRealtime
+DefaultGroupName=Voxtral Realtime
+OutputDir=installer-output
+OutputBaseFilename=VoxtralRealtime-Setup
+SetupIconFile=VoxtralRealtime\VoxtralRealtime\Resources\app.ico
+UninstallDisplayIcon={app}\VoxtralRealtime.exe
+Compression=lzma2
+SolidCompression=yes
+WizardStyle=modern
+ArchitecturesAllowed=x64compatible
+ArchitecturesInstallIn64BitMode=x64compatible
+PrivilegesRequired=lowest
+
+[Files]
+Source: "VoxtralRealtime\publish\*"; DestDir: "{app}"; Flags: ignoreversion recursesubdirs
+
+[Icons]
+Name: "{group}\Voxtral Realtime"; Filename: "{app}\VoxtralRealtime.exe"
+Name: "{autodesktop}\Voxtral Realtime"; Filename: "{app}\VoxtralRealtime.exe"; Tasks: desktopicon
+
+[Tasks]
+Name: "desktopicon"; Description: "Create a desktop shortcut"; GroupDescription: "Additional shortcuts:"
+
+[Run]
+Filename: "{app}\VoxtralRealtime.exe"; Description: "Launch Voxtral Realtime"; Flags: nowait postinstall skipifsilent