I Am Dictator 🎙️

Push-to-talk speech-to-text transcriber for Windows that lets you dictate text anywhere.

Features

• 🎯 Universal Hotkey: Press CTRL+ALT to start/stop recording
• 🤖 Local AI: Uses OpenAI Whisper for accurate transcription
• 🔇 Subtle Audio Feedback: Low-volume beeps indicate recording state
• ⚡ Batch Processing: Records complete utterances for better accuracy
• 🎤 Microphone Selection: Choose from available audio input devices
• 📝 Auto Text Injection: Automatically types transcribed text
• 🌍 Multi-language: Supports automatic language detection
• 🔒 Privacy-First: All processing happens locally

Quick Start

1. Run IAmDictator.exe
2. The app starts in the system tray
3. Hold CTRL+ALT and speak
4. Release keys when done
5. Transcribed text appears automatically

System Tray Menu

Right-click the tray icon to access:

• Language: Auto, English, Spanish, French, German, Portuguese
• Model: Tiny (fastest) to Large-v3 (most accurate)
• Injection Mode: SendInput or Clipboard
• Microphone: Select input device
• Settings: Advanced configuration
• Logs: View application logs

Requirements

• Windows 10/11
• .NET 8.0 Runtime
• Python 3.10+ (for STT server)
• Microphone

Architecture

┌─────────────────────┐
│   WPF Application   │
│  (System Tray UI)   │
└──────────┬──────────┘
           │
           ├── Hotkey Engine (CTRL+ALT)
           ├── Audio Capture (NAudio)
           ├── gRPC Client
           │
           v
┌─────────────────────┐
│   Python STT Server │
│  (faster-whisper)   │
└─────────────────────┘
           │
           v
┌─────────────────────┐
│   Text Injection    │
│   (SendInput API)   │
└─────────────────────┘

Components

• IAmDictator.App: WPF application with system tray UI
• IAmDictator.Core: Core functionality (hotkeys, audio, text injection)
• IAmDictator.STT.Server: Python gRPC server with faster-whisper

Configuration

Settings are stored in: %APPDATA%\IAmDictator\settings.json

{
  "Hotkey": {
    "Enabled": true,
    "ArmedDelayMs": 120
  },
  "Transcription": {
    "ModelName": "small",
    "Language": "auto"
  },
  "Injection": {
    "Mode": "SendInput"
  },
  "General": {
    "MicrophoneDeviceIndex": 0
  }
}

Building from Source

Prerequisites

• .NET 8.0 SDK
• Python 3.10+
• Visual Studio 2022 or VS Code

Build Steps

1. Clone the repository:

git clone https://github.com/yourusername/IAmDictator.git
cd IAmDictator

2. Build the solution:

dotnet build IAmDictator.sln --configuration Release

3. Setup Python STT server:

cd src/IAmDictator.STT.Server
python -m venv venv
venv\Scripts\activate
pip install -r requirements.txt

4. Run the application:

src\IAmDictator.App\bin\Release\net8.0-windows\IAmDictator.App.exe

Troubleshooting

No transcription appearing

• Check microphone selection in tray menu
• Verify STT server is running (check logs)
• Ensure correct microphone device is selected

Poor transcription quality

• Use a better microphone (headset recommended)
• Increase model size (small → medium → large)
• Speak clearly and at moderate pace
• Reduce background noise

Logs Location

%LOCALAPPDATA%\IAmDictator\logs\

License

MIT License - See LICENSE file for details

Credits

• Whisper: OpenAI
• faster-whisper: SYSTRAN
• NAudio: Mark Heath
• gRPC: Google

---

Made with ❤️ for productivity enthusiasts