SimpleWhisper

A Python GUI application for speech-to-text transcription using OpenAI's Whisper model.

Features

• Record audio from microphone
• Transcribe speech to text using Whisper models
• Automatic clipboard copying of transcribed text
• Audio device selection
• Configurable global hotkeys for system-wide recording control
• Local and global keyboard shortcuts
• Automatic temporary file cleanup
• Modern state-driven UI with progress indicators

Requirements

• Python 3.9+
• Audio input device (microphone)
• CUDA 12.8 Toolkit
• ffmpeg installation

Installation and Usage

1. Install UV (if not already installed):

   pip install uv

2. Clone or download the project:

   git clone <repository-url>
   cd SimpleWhisper

3. Install dependencies:

   cd src
   uv sync

4. Run the application:

   cd src
   uv run python main.py

Keyboard Shortcuts

Local Shortcuts (when app has focus)

• Ctrl+R - Start recording
• Ctrl+S - Stop recording
• Ctrl+T - Open settings

Global Shortcuts (configurable, work system-wide)

• Ctrl+Shift+R - Toggle recording (default)
• Ctrl+Shift+S - Stop recording (default)

Configure global hotkeys in Settings > Global Hotkeys tab.

Settings

• Audio Device: Select microphone input device
• Global Hotkeys: Configure system-wide keyboard shortcuts
• Whisper Model: Choose transcription model for accuracy vs speed

Settings are automatically saved to settings.json.