Workspace Launcher

Automatic work environment launcher for Windows. Launches configured apps, browsers and terminals, placing them on selected monitors — all with a single gesture.

Launch methods:

• Double clap (sound detection with PANNs CNN14 neural network)
• Voice command (offline speech recognition via Vosk)
• Keyboard shortcut (e.g. Win+Shift+W)
• System tray icon

---

Features

• Profiles — prepare different app sets (e.g. "Work", "Home", "Project X") and switch between them
• Window positioning — assign apps to specific monitors, screen halves (left/right) and layers (on top / behind)
• Smart launching — detects already running apps and just repositions them instead of launching again
• Sound recognition — PANNs CNN14 neural network classifies sounds (clapping, snapping, whistling, knocking and more)
• Voice command — offline, no data sent to the cloud (Vosk model)
• UWP support — automatically detects and launches Microsoft Store apps (Teams, Spotify etc.)
• Configuration GUI — graphical editor for profiles, apps and settings (customtkinter)
• Autostart — optional launch at Windows startup

---

Requirements

• Windows 10/11
• Python 3.10+ (to run from source)
• Microphone (for clap detection and voice commands)
• Optionally: Git Bash, AutoHotkey v2

---

Installation

Option A: From source (recommended for developers)

# Clone the repository
git clone https://github.com/YOUR-USER/workspace-launcher.git
cd workspace-launcher

# Install dependencies
pip install -r requirements.txt

# Run the configurator
python config_gui.py

# Run the launcher
python workspace.py

Option B: Compiled .exe

# Build .exe files
build.bat

# Output in dist/ folder:
#   WorkspaceLauncher.exe  — main application
#   workspace-config.json  — configuration

Copy dist/WorkspaceLauncher.exe + dist/workspace-config.json to any computer — no Python required.

---

Usage

1. Configuration (GUI)

Run the configurator:

python config_gui.py
# or
WorkspaceConfig.exe

In the configurator:

• "Apps" tab — add apps from the installed programs list or browse for .exe
• "Terminals" tab — add terminals (Git Bash, PowerShell, CMD, Windows Terminal) with startup commands
• "Settings" tab — set microphone sensitivity, keyboard shortcut, trigger sound, voice commands

For each app/terminal you can set:

Option	Description
Screen	Monitor number (1, 2, 3...)
Position	full, left half, right half
Layer	Normal, On top, Behind
Order	Launch priority (higher = earlier)
Minimize	Launch minimized

2. Running the launcher

python workspace.py
# or
WorkspaceLauncher.exe

The application:

1. Starts in the system tray
2. Loads the sound recognition model (~80 MB, one-time)
3. Listens on the microphone — displays volume bar in the console
4. After detecting a double clap (or hotkey/voice command) launches the active profile

3. Triggering workspace

Method	Default setting
Clap	2x hand clap
Hotkey	Win+Shift+W
Voice command	Set in configuration (e.g. "launch")
Tray icon	Right-click > "Launch: [profile]"

4. Tray menu

Right-click the tray icon:

• Launch: [profile] — launch selected profile
• Close workspace — close all launched processes
• Configuration — open configuration GUI
• Quit — close Workspace Launcher

---

Configuration (JSON)

The workspace-config.json file is created automatically by the GUI. Example configuration:

{
  "czulosc_klasniecia": 70,
  "hotkey": "Win+Shift+W",
  "zdarzenie_dzwiekowe": "Clapping",
  "liczba_zdarzen": 2,
  "cooldown": 3,
  "czulosc_nn": 0.12,
  "slowa_kluczowe": "launch",
  "jezyk_mowy": "en",
  "profil_aktywny": "Work",
  "profile": {
    "Work": {
      "aplikacje": [
        {
          "nazwa": "VS Code",
          "exe": "code",
          "argumenty": "C:\\Projects\\my-app",
          "ekran": 1,
          "polowa": "",
          "warstwa": "Normal",
          "kolejnosc": 0,
          "minimalizuj": false
        }
      ],
      "terminale": [
        {
          "nazwa": "Dev Server",
          "terminal_typ": "Git Bash",
          "folder": "C:\\Projects\\my-app",
          "komenda": "npm run dev",
          "ekran": 2,
          "polowa": "right",
          "warstwa": "On top"
        }
      ]
    }
  }
}

Full example in workspace-config.example.json.

Configuration options

Key	Type	Description
czulosc_klasniecia	int (40-95)	Microphone volume threshold in dB
hotkey	string	Keyboard shortcut, e.g. Win+Shift+W, Ctrl+Alt+S
zdarzenie_dzwiekowe	string	Sound type: Clapping, Finger snapping, Whistling, Knock, Bell etc.
liczba_zdarzen	int (1-3)	How many times to repeat the sound to trigger
cooldown	int (0-30)	Pause between launches in seconds
czulosc_nn	float (0.05-0.50)	Neural network confidence threshold (lower = more sensitive)
slowa_kluczowe	string	Voice commands separated by commas (empty = disabled)
jezyk_mowy	string	Speech recognition language: pl or en

---

Additional scripts

File	Description
clap-trigger.py	Standalone clap trigger (simple, no neural network)
voice-trigger.py	Standalone voice trigger (Google Speech Recognition)
workspace-launcher.ps1	PowerShell launcher (independent of Python)
workspace-hotkey.ahk	AutoHotkey v2 shortcuts (Ctrl+Alt+W, Ctrl+Alt+P)
create-shortcut.vbs	Creates Windows startup shortcut
build.bat	Compiles .exe using PyInstaller

Standalone clap trigger

# Listen for double clap
python clap-trigger.py

# Calibration mode — check microphone level
python clap-trigger.py --calibrate

# Change threshold and profile
python clap-trigger.py --threshold 65 --profile praca --debug

---

Building .exe

# Install PyInstaller
pip install pyinstaller

# Build WorkspaceLauncher.exe (console)
pyinstaller --onefile --name WorkspaceLauncher --console workspace.py

# Build WorkspaceConfig.exe (no console)
pyinstaller --onefile --name WorkspaceConfig --windowed config_gui.py

Or use the ready-made script:

build.bat

---

Project structure

workspace-launcher/
├── workspace.py              # Main application (tray + detection + launcher)
├── config_gui.py             # Configuration GUI
├── clap-trigger.py           # Standalone clap trigger
├── voice-trigger.py          # Standalone voice trigger
├── workspace-launcher.ps1    # PowerShell launcher
├── workspace-hotkey.ahk      # AutoHotkey v2 shortcuts
├── workspace-config.json     # Configuration (created by GUI)
├── workspace-config.example.json  # Example configuration
├── build.bat                 # Build script for .exe
├── create-shortcut.vbs       # Startup shortcut creation
├── create-startup-shortcut.ps1
├── clap-trigger-startup.bat
├── requirements.txt          # Python dependencies
└── .gitignore

---

License

MIT