VOX – Desktop Voice Assistant

VOX is a Python-based desktop voice assistant that enables hands-free system interaction using a custom wake word. It supports real-time voice commands, system automation, and natural voice responses through a lightweight floating widget interface.

Features

• Wake-word detection using Picovoice Porcupine (“Hey Vox”)
• Real-time speech recognition
• Hybrid text-to-speech (Edge TTS with pyttsx3 offline fallback)
• Floating, draggable widget UI using CustomTkinter
• System automation (open apps, system settings, volume control, sleep mode)
• Clipboard actions (copy, paste, save, reload)
• Voice-based Google and YouTube search
• Task management using voice commands with JSON storage
• Contextual responses such as time, weather, and greetings
• Audio feedback for command success and failure
• Multithreaded and asynchronous execution for smooth performance

Tech Stack

• Language: Python
• UI: CustomTkinter, Tkinter
• Wake Word: Picovoice Porcupine
• Speech Recognition: SpeechRecognition
• Text-to-Speech: Edge TTS, pyttsx3
• Audio: PyAudio, Pygame
• Automation: PyAutoGUI, ctypes
• Storage: JSON

Installation

git clone https://github.com/Midhun1618/VOX.git
cd VOX
pip install -r requirements.txt

Note: PyAudio may require a pre-built wheel on Windows.

Run

python main.py

Example Commands

• Hey Vox, open YouTube
• Hey Vox, what is Python multithreading
• Hey Vox, add task buy groceries
• Hey Vox, read my tasks
• Hey Vox, mute volume

Project Structure

VOX/
├── assets/
├── main.py
├── tasks.json
└── README.md

Author

Midhun KP
GitHub: https://github.com/Midhun1618
LinkedIn: https://www.linkedin.com/in/midhun-kp