Inspiration
Hello everyone, this is Gauraw, and I built ARIA AI for the MEDO competition.
I wanted to create something that goes beyond simple grammar correction. Many people write every day, but they often struggle to understand why mistakes happen or which areas need improvement. I wanted to build a system that analyzes writing patterns, tracks progress over time, and helps users improve communication in a more structured way.
What it does
ARIA AI analyzes writing and identifies weak areas in sentence structure, grammar, and overall communication.
Key features include:
- Tense analysis
- Analytics dashboard
- Speech-to-text support
- Direct text input
- Calendar integration
- Mistakes tracking tab
- To-do system
- Progress tracking with Supabase database
- Optional local saving mode for users with privacy concerns (only Gmail ID sync is used)
Instead of only correcting text, ARIA aims to help users understand and improve their writing habits over time.
How I built it
I used Supabase for storing and syncing user progress and data. The application combines text analysis, speech-to-text functionality, analytics, and productivity tools into one ecosystem.
The goal was to make the experience simple while also giving users useful insights about their communication patterns.
Challenges I ran into
One of the biggest challenges was balancing features with simplicity. Integrating multiple systems such as analytics, speech-to-text, tracking, and data synchronization while keeping the user experience smooth required a lot of iteration.
Another challenge was handling privacy concerns. To address this, I added an optional local-saving mode so users can choose how they want their information stored.
What I learned
This project taught me a lot about building AI-powered experiences that are useful in real-world situations. I also learned the importance of user feedback, privacy choices, and creating features that solve actual problems rather than adding complexity.
Future plans
I’m very open to feedback and improvements. I plan to continue refining ARIA AI by adding better analysis systems, improving accuracy, and expanding features based on community suggestions.
Every suggestion helps make ARIA better.
Built With
- ai-text-analysis
- authentication
- calendar
- local-storage
- postgresql
- react
- speech-to-text-apis
- supabase
- tailwind-css
- typescript
- vite
Log in or sign up for Devpost to join the conversation.