How can I make real-time sign language recognition system?

Hi Guys,

I want to make a real-time sign language detection system where a user will be doing these gestures in front of the camera and get the corresponding letters on the screen. And then form the words and sentences.

For now, I am using this American sign language dataset available on Kaggle. Here is the dataset link ASL Alphabet | Kaggle

I have trained my model but I don’t have any idea about making such a system.

Can anybody point out me to any kind of resources like Github repo, articles, tools, frameworks anything or if you guys have done something like this before please write your thoughts and implementation ways on this post

Hey @RadhikaBansal I implemented the exact thing you are asking for sometimes back
Check it out in these blog posts I wrote:
Part One: Training the Neural Network
Part Two: Creating the Real time System

Its a start and you can improve from there. Hope that helps


Hyy @jimmiemunyi !!! I went through your blogs and these are awesome. You have explained each and everything so well. I thought it was a difficult task but these blogs are great help. .

Thank you!!


I started learning them a few days ago from a YouTube tutorial. In this regard, you have two options one could be making them by using the Python framework. The second option could be its detection by using Tensorflow Object Detection.