Inspiration
Students are constantly stuck choosing between listening and taking notes—and usually, they do both poorly. In fast-paced lectures, key ideas get missed, notes become incomplete, and understanding suffers. We wanted to eliminate that trade-off entirely. The idea behind NEXA was simple: what if students could focus 100% on learning while an intelligent system captured and organized everything for them in real time?
What it does
NEXA is an intelligent lecture companion that records live audio from online or physical classes and converts it into real-time transcripts. But it doesn’t stop there—it uses AI to transform raw speech into structured notes, including key points, summaries, and important terms. The result is a clean, organized learning output that students can immediately review and use.
How we built it
We built NEXA as a lightweight browser-based application using React and Tailwind CSS for the frontend. For real-time transcription, we integrated the Chrome Speech Recognition API, enabling continuous speech-to-text streaming. A service worker handles background processing, receiving transcript chunks and generating insights using AI. The UI was designed to update dynamically, ensuring a smooth, real-time experience.
Challenges we ran into
One of the biggest challenges was handling real-time data flow without lag. Streaming speech, updating transcripts, and generating AI insights simultaneously required careful state management. Another challenge was maintaining a responsive UI while processing continuous input. Synchronizing transcript updates with AI-generated summaries without breaking the user experience was a critical hurdle.
Accomplishments that we're proud of
We successfully built a system that feels alive—transcripts update instantly, and AI insights appear in real time without interrupting the user. The clean, intuitive UI allows users to interact with complex functionality effortlessly. Most importantly, we created a working solution that directly addresses a real problem students face every day.
What we learned
We learned how to design and build real-time systems that balance performance and usability. Working with streaming data taught us the importance of efficient state management and asynchronous processing. We also gained deeper insight into how AI can be practically applied to enhance productivity, not just automate tasks.
What's next for NEXA
Next, we plan to improve transcription accuracy using more advanced models and introduce speaker detection. We also want to add cloud syncing, so users can access their notes across devices. Future updates will include flashcard generation, smart highlights, and deeper AI-powered study tools. Ultimately, we aim to turn NEXA into a complete learning assistant that goes beyond note-taking and actively helps students understand and retain knowledge.
Log in or sign up for Devpost to join the conversation.