AutoScribe

Inspiration

We noticed that in many meetings, people get lost in technical terms and long explanations. Misunderstandings pile up, and important ideas get lost. We wanted to make meetings easier for everyone by turning spoken ideas into clear visuals instantly.

What it does

Autoscribe listens to what you say and automatically creates diagrams that represent your ideas. It transforms complex talk into simple, easy-to-follow visuals, making sure everyone is on the same page.

How we built it

We combined speech to text technology, by using Assembly AI API (in real time). It understands the natural language. Then, we generate a prompt and used Open AI api to create Mermaid.js syntax, stored it in our database (Supabase) and render it in frontend (React.tsx and TailwindCSS).

Challenges we ran into

Keeping the system fast enough so the diagrams update in real time.
Understanding how openAI works and how to integrate it in our system.

Accomplishments that we're proud of

Learn real-time speech to text
Achieving real-time updates that actually keep up with the conversation.
Able to work effectively together as a team.

What we learned

We learned how powerful visuals can be in communication. We also discovered that combining speech recognition with diagram generation is harder than it looks but extremely valuable. Most importantly, we learned that technology can really bridge the gap between experts and non-experts in a conversation.