Inspiration

Two people in our group have been in Canada for less than 3 years. Phineas told us a story of how his grandparents who do not speak English would narrate long stories about their life. After that, he would write all this information down and organize it to fill in visa forms, health cards, and other legal documents. We found this process way too time-consuming and inefficient. Legal Lens was built exactly to solve this problem.

What it does

Legal Lens performs image-to-text transcriptions for legal documents from over 100+ languages available, translates them to English, and organizes them into straightforward, key-value format.

How we built it

We built this project in plain html/css/javascript and used Tesseract.js for image-to-text transcription, EdenAI for translating text, and ChatGPT API for organizing and processing text to send back to the user.

Challenges we ran into

One of the biggest challenges we've faced was implementing OpenAI's gpt model APIs as we needed to figure out if there were other better-suited alternatives out there for a similar language processing API to return the parameterized text.

Accomplishments that we're proud of

We were capable of creating something truly powerful and useful; We can see Legal Lens growing and developing into something bigger, and we're really proud with the result of our very first hackathon.

What we learned

We learned how to implement and use various APIs in coordination with each other, and more importantly, how to work in a group and manage tasks in an effective way that would highlight on each member's strengths.

What's next for Legal Lens

We are considering adding voice detection to our text box, though we already had a pretty good solution for voice dictation using the text box. We also thought of creating a web extension/browser plugin that would save that information and help the user create personal info auto-fills for any online government forms.

Built With

+ 5 more
Share this project:

Updates