Inspiration
The world is becoming increasingly digital, but accessibility challenges remain a major hurdle for people with disabilities. From visual impairments to neurodivergence and motor disabilities, many users face barriers that hinder their ability to perform simple online tasks. Inspired by these challenges, we sought to create a solution that uses semantic search, AI, and accessibility-first principles to enhance the online experience for everyone.
This project tackles real-world challenges in accessibility by leveraging MongoDB Atlas and AWS services to improve usability for individuals with disabilities while demonstrating the power of combining vector databases with generative AI.
What it Does
Our application is an AI-powered accessible search platform designed to make online shopping and browsing more inclusive. The platform enhances the user experience through these core features:
- Natural Language Query Search:
Users can input conversational queries (e.g., “Show me comfortable shoes for winter”). This is especially helpful for individuals relying on screen readers or voice input.
- ##Error Tolerance:
Misspellings, synonyms, and paraphrased queries are automatically corrected and interpreted. For example, a search for “jens” will return results for “jeans.”
- ##Voice Search Integration:
Designed with motor disabilities in mind, the app allows users to perform searches via voice commands, offering a hands-free experience.
- ##Personalized Recommendations:
The app offers tailored product suggestions by analyzing user preferences and context, simplifying decision-making for users with cognitive disabilities.
- ##Accessible Product Descriptions:
Semantic AI generates simplified descriptions or categorizes products for accessibility (e.g., “makeup for sensitive skin”).
- ##Enhanced Image Search:
Powered by image recognition, users can describe visual elements in natural language, such as “blue dress with floral patterns.”
Log in or sign up for Devpost to join the conversation.