Signify

Inspiration

We recognized that the deaf and mute community are underserved in the technology sphere

What it does

Translates ASL into English Text, that can also be read aloud.

How we built it

We used teachable machine to create a TensorFlow model. The user interface was built using JavaScript specifically p5.js.

Challenges we ran into

Whilst creating the model we ran into many problems like runtime, skin tone discrepancies, and collecting a large enough dataset that could be trained within the time limit.

Accomplishments that we're proud of

We are proud of creating our first object detection machine learning model.

What we learned

To make the dataset we learned image manipulation, and for the frontend we learned p5.js specifically webcam access as well as linear and radial gradients. We also learned the ASL alphabet.