An Indian engineering student has developed an innovative AI model that can translate American Sign Language (ASL) into English words.

An Indian engineering student has developed an innovative AI model that can translate American Sign Language (ASL) into English words.

The model can recognize and translate six fundamental ASL signs like ‘Hello’, ‘I Love You’, ‘Thank You’, ‘Please’, ‘Yes’, and ‘No’.

The AI model, developed by Priyanjali Gupta, a student of Vellore Institute of Technology, was built using TensorFlow Object Detection API, utilizing transfer learning with the ssd_mobilenet model.

Gupta captured ASL gestures through a webcam that formed the basis for her AI model. In response to inquiries, Gupta acknowledged the complexity of building a dedicated deep learning model for sign detection. 

As part of a comment on a LinkedIn post, Gupta stated “You are absolutely right there’s a long long way to go to make a perfect model which can be implemented in real life. Hopefully this happens in our lifetime”.

Gupta’s AI model helps people communicate across barriers, providing a more accessible way to connect with the world. In her GitHub post, she has shared her project and code.

source/comments: thehindu.com (headline edited)