Accessible Digital Control System — Empowering hands-free interaction through voice commands and head gestures.
Unbound is a hands-free digital control platform that allows users to operate a computer using only voice commands and head gestures.
Built to bridge the accessibility gap, it empowers individuals with motor impairments, upper limb disabilities, temporary injuries, or fatigue to navigate, communicate, and interact — independently and confidently.
- Open apps and websites (Google, YouTube, Stack Overflow, etc.)
- Play YouTube songs via voice
- Send WhatsApp messages
- Send emails
- Fetch weather and news updates
- Set alarms and reminders
- Adjust system volume and brightness
- Scroll, navigate, and automate system tasks
- Engage in AI-based conversations (using OpenAI GPT)
- Translate text between languages (e.g., English to Hindi, French, etc.)
- Fun utilities: Toss coin, roll dice, tell jokes, fun facts
- Hands-free mouse navigation via voice
- Move mouse pointer using simple head movements
- Perform clicks without any physical device
- Built using only a webcam — no external hardware needed
- Optimized for smooth, fatigue-free control
Prototype Videos:
- Voice Mode: Command-driven task execution
- Gesture Mode: Cursor control with head movements
- [Add YouTube or Drive Link Here if needed]
- Python 3.9+
- SpeechRecognition — Voice input processing
- PyAudio — Microphone handling
- pyttsx3 — Text-to-Speech (TTS) output
- PyAutoGUI — Mouse, scroll, and keyboard automation
- OpenCV — Real-time head tracking
- TensorFlow Lite — Lightweight models for gesture interpretation
- Requests — API requests for weather and news
- PyWhatKit — WhatsApp messaging, YouTube control
- Googletrans — Translation services
- OpenAI GPT API — AI chat assistant
- Minimal Setup: Just a webcam and a microphone.
- Real-Time Performance: No lag, no bulky hardware.
- Designed for Dignity: Not just access, but independence and freedom.
- Expandable Foundation: Ready for future modules like predictive navigation and personalized learning.
- 🔗 Merge voice and gesture input seamlessly for hybrid control
- 🧠 Add adaptive learning for custom gestures and voice styles
- 🎯 Improve personalization (predictive shortcuts, recommendations)
- 🦮 Accessibility upgrades (blind-friendly audio navigation)
- Namit Rana -
- Khushboo Kataria -
Built with ❤️ at Geekverse IEEE GGSIPU 2024.