This is a submission for the Google AI Studio Multimodal Challenge
What I Built
I built an AI-powered lecture assistant that transforms lengthy lecture transcripts into concise summaries, generates questions for self-assessment, and even crafts content ideas for blogs or learning journals on platforms like Medium, Dev.to, or personal blogs.
This web app operates in two phases:
Lecture Analysis & Summarization:
- Accepts raw lecture text input
- Sends it to a custom AI workflow built in Google AI Studio
- Instantly returns a clean, easy-to-digest summary
Content & Learning Assistance:
- Generates self-assessment questions based on the lecture
- Provides article and content ideas so learners can document their study journey or share insights online It’s a lightweight, accessible tool designed for students and content creators who want to learn, review, and share knowledge efficiently.
Demo
Originally, I planned to deploy the app using Google Cloud Run, but due to card verification issues, I switched to Netlify for hosting so reviewers can still access the live demo seamlessly.
- Live Demo (Hosted on Netlify): Netlify
- GitHub Repository: Github repo
- AI Studio Workflow File: View Saved File
And also here is the link to the full demo video on Youtube.
Video link
Screenshots:
here are the screen shot of some of the ui:
this is the landing page:
here is the search box and the result:
How I Used Google AI Studio
I used Google AI Studio to:
- Design and test the AI prompt for summarizing lecture transcripts
- Fine-tune the response formatting for clarity and conciseness
- Export the workflow so it could be integrated into a simple web application
The saved AI Studio file above verifies the use of the platform and allows others to see exactly how the workflow was built.
Multimodal Features
Our system transforms lecture content into interactive, multimodal learning experiences through the following stages:
Audio/Video → Text → Structured Modules
Gemini-2.5-Flash converts lectures into accurate transcripts, then organizes them into learning modules with quizzes, flashcards, and AI-generated practice questions.Text → Image Storytelling
Imagen-4.0-Generate-001 brings story-driven learning segments to life as engaging visuals, creating strong memory cues for students.Text ↔ Image Matching Challenges
Learners complete interactive matching exercises before lessons begin—pairing terms with images to build active recall and conceptual understanding.Text → Speech Narration
Using the Web Speech API, summaries and story scenes are turned into clear, natural-sounding audio to support auditory learners and improve accessibility.
Conclusion
This project demonstrates how Google AI Studio can be combined with a simple frontend to deliver practical, student-focused solutions in a short amount of time.
All source code, demo links, and workflow files are available above for anyone to explore or extend.
Top comments (0)