Inspiration
In the covid situation, we observed that many meetings and discussions were online where deaf people face issues while communicating with us. We couldn't understand their sign language, so we are planning to provide a platform for them and us as well to communicate easily without any Interpreter.
What it does
It converts sign language into human-readable language
How we built it
We build using Machine Learning with CNN Model with 85% Accuracy and we are also able to detect alphabets live.
Challenges we ran into
We are facing troubles when we are doing live detection and another issue is the lack of data to train
Accomplishments that we're proud of
We are able to achieve recognition of Sign Language Alphabet with 85% Accuracy.
What we learned
We learned about, how real-time detection has much more complications than static image detection.
What's next for American Sign Language Detection Using CNN
We are planning to detect expression/words in Real-Time with maximum reliable accuracy and speed.
Log in or sign up for Devpost to join the conversation.