Inspiration
One of the teams members has a family memeber that is deaf.
What it does
It uses machine learning to translate finger positions to the alphabet
How we built it
We built it using Python and HTML. We used Opencv, Mediapipe and TensorFlow.
Challenges we ran into
The software requires near perfect accuracy to the dataset images, so it has made replicating signs and converting them to letters hard
Accomplishments that we're proud of
Being able to stream it the the website and display translation
What we learned
We learned how machine learning works and how it trains off of data
What's next for OctoLingo - Sign Language Translation Project
An transformation into an overlay software so that people can use it in any video calling application; as well as, image-to-text, and text-to-speech.
Log in or sign up for Devpost to join the conversation.