write:here

Say goodbye to typing, and hello to quick and easy digitalization of handwritten letters! Write:here improves accessibility with a simple, user-friendly interface that transforms letters into emails.

Comment

Inspiration

Not all of us had the privilege to grow up with a keyboard at our fingertips, but that shouldn't prevent anyone from staying connected in today's digital world. We all know someone in our life who prefers the simplicity of a handwritten letter over navigating the complex mailing system just to enter each character one key at a time. This person may be a grandparent, a parent, a friend, or even yourself. We wanted to break down these accessibility barriers by providing individuals with the option to send emails from handwritten letters.

What it does

Write:here scans an image file for handwritten text and parses it into digital text using AI-powered technology. It allows the user to send the text as an email through our mailing platform by providing the recipient's email address as well as their own.

How we built it

In the initial stages of development, our team utilized Figma to develop a comprehensive mockup of the project, providing a clear representation of its intended design and functionality. Building upon this foundation, we used Next.js to construct the front and back end of the project, integrating back-end functions with their respective front-end components. Through page routing, we were successful in creating a fully-functional, integrated application."

Challenges we ran into

Our team encountered several challenges during the development process, as this was the first hackathon for three out of four members. Additionally, our team had limited experience with Next.js, which presented an obstacle in itself. We also faced difficulties in implementing AI, such as Google Cloud Vision due to the lack of available documentation and co:here using prompt engineering. The deployment phase also presented its own set of challenges, including the use of outdated libraries and extensive peer dependencies. Despite these obstacles, our team persevered and was able to overcome these challenges, ultimately delivering a successful project.

Accomplishments that we're proud of

  • We were able to use Google Cloud Vision to scan and parse image files with text.
  • Web Capture
  • Learning how to use Co:here and creating 2 work-in-progress features using co:here
  • Submitting a project in our first hackathon!
  • Typeform-like animated form

What we learned

  • Effective prompt engineering
  • Training a model
  • API implementation
  • Next.js : this was brand new to 3/4 of our team members!
  • Project deployment
  • Collaborating using Github

What's next for write:here

  • Implement subject line generator and email formatter
  • Deploy as a mobile app
  • Gathering feedback from user testing

Built With

Share this project:

Updates