Inspiration
As an undergraduate, I often upload lecture notes to generative AI tools like ChatGPT to explain concepts, create notes, multiple-choice questions for practice, and use Gemini create audio overviews of the lecture material. However, I started wondering what it would be like to have a single, well-optimized platform to achieve all these tasks. That is why I developed UniBuddyAI.
What it does
It simplifies the learning process and makes it neither boring nor tedious.
How we built it
I utilized the Bolt.new platform for approximately 98% my project work, making only minor adjustments. For backend services, I used Supabase, the AI models GPT-3.5 and GPT-4 were used. Audio was generated using Eleven Labs.
Challenges we ran into
The primary challenge was the OCR process. Lecture notes often contain diagrams tables, and images, which made converting them into a readable format particularly difficult.
Accomplishments that we're proud of
I am so proud of building such a tool within a short period of time using the Bolt platform.
What we learned
I have learned a lot of new:
- How to a quick prototype
- How to process using Azure Document Service
- How to generate audio with ElevenLabs
- Use Supabase storage
- Write edge functions in Supabase
and more.
What's next for UniBuddyAI
I intend to further optimize this platform and introduce it to the market as a product targeted at under.
Log in or sign up for Devpost to join the conversation.