Jargon

Welcome to Jargon!

Inspiration

While we spent much of our hackathon deliberating between ideas and across verticals, we eventually decided on innovating within the relatively stale space of language learning. There are many software services that teach you languages, but they all follow a hard-coded hierarchy. There is no supplement to what you learn, and it's easy to lose interest as you learn "apple" and "it's raining" before "how do I get to the nearest train station?" or "what's some advice you have for someone looking to work overseas?".

It has long been shown that immersion is the BEST way of learning a language. So, rather than centralizing language learning within an app, we decided to make the user's learning experience a universal one. A model of learning where you're forced to actively recall, make decisions, evaluate context, and more. And in a setting where you're guaranteed to be invested in what you're reading.

So came about Jargon.

What it does

Jargon is a chrome extension that utilizes optical character recognition and large language models to universally recognize, translate, and quiz a user about text on ANY webpage. Imagine you're scrolling through Reddit or Facebook, CNN or The Guardian. Every so often, a popup will take a sentence from the article, comment, or caption and translate it into a language of your choice with one word missing. Your job is to select the right choice amongst a set of carefully generated (real) alternatives! Toggle language, proficiency, query frequency, and more with our plugin!

How we built it

Jargon utilizes OpenAI's GPT-4 and Amazon's Textract to scan and analyze, then translate, any webpage. A Plasmo-based Chrome extension runs both APIs and provides a pleasant UI/UX.

Challenges we ran into

We encountered many challenges interfacing with our APIs as well as getting them to integrate with the modern chrome landscape. We were forced to code and troubleshoot deprecated software multiple times before realizing that we had to move to another framework. Ultimately, we were able to succeed and learned the dangers of not properly planning out our development cycle.