The developed application is an innovative solution aimed at assisting visually impaired users in navigating their daily tasks through a combination of advanced technologies, including Artificial Intelligence (AI), voice commands, and gesture-based controls. The primary focus of the app is to enhance accessibility for blind or visually impaired individuals by providing real-time object recognition, text scanning, safe navigation features and OpenAI functionalities.
- Swift [SwiftUI + UIKit]
- Xcode 15
- Apple ML
- Apple VisionKit
- MongoDB Realm
- Accessibility features
- OpenAI API
- Ampel-Pilot-iOS
- YOLOv3
- Lottie
- Swinject
- Text-to-speech (TTS) and Speech-to-text (STT)
Finished
Apple iPhone with iOS 15+ installed
- Pedestrian crossing light color indicator: Identify the color of pedestrian crossing's light to provide user with real-time navigation suggestions. Powered by Ampel-Pilot-iOS
- Real-time Object Recognition: By using the device’s camera, the application identifies objects and obstacles in real-time, ensuring users can avoid hazards during movement. It also alerts the user through vibrations and sound signals. Powered by YOLOv3
- Text Scanning and Voice Feedback: The app allows scanning of textual documents or product labels and reads the content aloud using text-to-speech (TTS) technology.
- Color identification: Use the camera to identify color of dominant object in the area or measure
- Light intensity measurer: Use the camera to measure light intensity and notify the user by playing the sound which volume increases as light intensity level increases
- OpenAI API Interaction: The integration of AI models (such as GPT) enables the app to handle complex queries and provide dynamic responses, offering assistance to users by answering their question or analyzing photos.
- Gesture and Voice Commands: To ensure ease of use, the app supports various custom touch gestures (e.g., tap, swipe) and voice commands (speech-to-text STT technology), enabling users to perform tasks hands-free, reducing the need for tactile input.
- Customizable Interface: Users can adjust the app’s speech speed, voice, and even the interface language to suit their individual preferences.
- App Guide: Get to know about all the app's functionalities by completing the onboarding which teaches the user how to use the app
