Inspiration
One of our cousins works in teaching ASL and we wanted to create something that could be helpful.
What it does
ASLingo is using a neural network to recognize ASL. It uses a GUI which allows you to generate a random word and then you have to spell it out. The AI will then check to make sure you are spelling it correctly and tell you if your correct at the end.
How we built it
We built this with python. The libraries that we used were tensor flow for the neural network and PyQt5 for the GUI.
Challenges we ran into
Our original plan was to host the AI on a server so that ASLingo could be accessed on a website but it was very laggy so instead we opted to create a GUI in python.
Accomplishments that we're proud of
We are very proud of gathering our own data to train the neural network on and creating the neural network in general as it was our first time.
What we learned
We learned a lot about gathering data and training neural networks as well as creating GUI's.
What's next for ASLingo
To further build on ASLingo we would like to add lessons to turn this into an educational platform.
Log in or sign up for Devpost to join the conversation.