Project Story: VisionVoice
Inspiration
The inspiration for VisionVoice came from a heartfelt conversation with a visually impaired friend who expressed the difficulties of missing out on non-verbal cues during interactions. This motivated us to leverage AI technology to create a tool that empowers the blind community, making communication more inclusive and meaningful.
What it does
VisionVoice is an AI-powered application that helps blind individuals experience visual communication through descriptive audio or vibration feedback. The app interprets and conveys the expressions, reactions, and activities of their conversation partners, enabling more enriching and effective social interactions.
How we built it
Research and Planning: We conducted extensive research on the communication challenges faced by the blind community, outlining the key features and functionalities required for VisionVoice.
Design and Prototyping: We designed user-centric interfaces with accessibility in mind, ensuring ease of navigation. Prototyping helped us visualize the user flow and overall interface.
Development: Using AI and computer vision technologies, we developed algorithms to interpret facial expressions and body language. These were integrated with text-to-speech and vibration feedback mechanisms.
Challenges we ran into
Technical Complexity: Developing accurate AI models to interpret expressions and actions required extensive training data and fine-tuning to achieve reliable performance.
Accessibility Design: Ensuring the app was fully accessible and intuitive for blind users demanded meticulous attention to design details and continuous user testing.
Feedback Mechanism: Creating a seamless and effective feedback system that conveyed detailed information through audio and vibration without overwhelming the user was a complex task.
Accomplishments that we're proud of
We are proud of developing a functional and intuitive application that addresses a significant need in the blind community. The positive feedback from our users has been incredibly rewarding, affirming the impact of our efforts.
What we learned
Through this project, we gained a deep understanding of the needs and experiences of visually impaired individuals. We learned the importance of intuitive design and accessibility in app development, and we expanded our knowledge of AI and computer vision technologies.
What's next for VisionVoice
We plan to further refine VisionVoice based on user feedback, improving its accuracy and functionality. Our goal is to expand the app's capabilities, incorporating more nuanced interpretations and exploring additional feedback methods to better serve our users.
Built With
- coreml
- python
- swiftui
Log in or sign up for Devpost to join the conversation.