🧭 Sprint 3 – AI and Navigation Enhancements

🎯 Sprint Goal

The primary goal of Sprint 3 is to significantly enhance the accessibility and usability of the Mind-For-The-Blind application by introducing voice-based interactions, improving the accuracy of the ML model, and refining the overall user experience through thorough testing and UI adjustments.

---

✅ Objectives

1. Integrate Voice-Based AI Assistant

• Add a conversational AI assistant for handling user queries through voice input.
• Enable speech-to-text and text-to-speech functionalities for seamless communication.

2. Improve Voice Accessibility & App Guidance

• Implement audio prompts on screen load to assist blind users.
• Ensure all critical pages provide spoken instructions and context.

3. Improve ML Model Accuracy

• Retrain or fine-tune the currency detection ML model.
• Enhance accuracy and reduce latency for real-time usage.

4. Enable Realtime Camera Access for Currency Scanning

• Provide a responsive and low-latency camera preview.
• Ensure real-time detection and audio announcement of dollar denominations.

5. UI Refinements & Accessibility Improvements

• Improve layout, text contrast, and element spacing for accessibility.
• Ensure compatibility with screen readers and improve user flow.

6. Conduct Broader User Testing & Collect Feedback

• Involve both blind users and helpers for hands-on testing.
• Document feedback and identify areas for improvement.

7. Resolve errors that we are encountering in the Video call module

• Try integrating the ZEGOCLOUD module to see if we can successfully integrate the module.
• If not find alternative ways.

---