We wanted to help people maximize their time and make studying more accessible.
What it does
SnapStudy is a Chrome extension that allows it users summarize their active webpage, option to Read Aloud the summary and download it as a pdf using Gemini AI. In addition, it has a options page where users can put in their preferences i.e., I want summary in Cornell Notes, etc. and their preferred languages and saves them for future use.
How we built it
We started by building a simple Chrome extension using Manifest V3 to get an idea of it works. Then after some research, we got the extension to scrape the text of the active webpage using JavaScript functions. Next, we made it to send the scraped text to the Gemini Model using its API with a Post Method obtained from Gemini documentation. We then added an options page allowing users to customize the system instructions sent to Gemini. Then we used built-in JS libraries to add a Read Aloud function to read the summary out loud. Then we added language options to the user options page so users can select their preferred text language and voice assistant. Lastly, we used CSS libraries to enhance the UI and make it engaging and aesthetic looking.
Challenges we ran into
First challenge was extracting text from the webpages.
Second problem was saving and retrieving the options saved by the user on the options page.
Third problem was UI formatting, mainly centering elements and text.
Accomplishments that we're proud of
We made significant progress on our project competing in any Hackathon for the first time.
We learned how to make a Chrome extension from scratch.
We successfully integrated the Gemini model in our project using API keys.
We used HTML and CSS to create a aesthetic looking UI.
What we learned
Making a Chrome Extension.
How to use APIs to send prompts to a LLM (Gemini).
How to market our product i.e. creating a logo, taskbar icon, Names, etc.
What's next for SnapStudy
Some ideas we have include:
Real time text magnification: allow users to hover over text to enlarge it dynamically
Adjustable speed and voice pitch
Sign language video support: integrate AI based sign language translation for the text
AI Chatbot for easy navigation: a chatbot that guides users through the page and asks questions
Voice command support: allows users to control reading and summarization via voice commands
Log in or sign up for Devpost to join the conversation.