Inspiration
To help the specially challenged
What it does
Help the specially challenged to express their ideas to more people
How we built it
using python, and the tensorflow, opencv, and cvzone modules
Challenges we ran into
The data collection was a hard part, since it was hard to get more variability in the data with just the two of us. This also impacted our model accuracy
Accomplishments that we're proud of
Making a final working model with OpenCV is something we did for the first time. We are very proud of the solution we finally came up with, and believe that given more time and effort, we can transform it into software with revolutionary capabilities.
What we learned
We learnt the importance of having a never-give-up attitude. The task seemed very intimidating to us, given the limited time and members on our team. We spent countless hours working on getting the model up and working, and we are happy with the end result since we believe it is reflective of our attitude. No matter we win the hackathon or not, we are extremely proud to have make the progress we did
What's next for Sign Language Interpreter
next up, we want to expand the signs it is capable of detecting, and also add a mode where the speaker can simply speak in alphabets, and the letters are saved in a file and displayed to the listener. We would love to improve the model by getting data from a variety of people.
Log in or sign up for Devpost to join the conversation.