Wow! This Indian student did what NO big tech firm could! Real-time translation of sign language

Priyanjali Gupta, the third-year Engineering student was motivated to do something out of the box. No, not with some inspiration but due to her mother who used to ask her to do something while studying. Well, that's enough to create the magic! And she did it.. The engineering student at Vellore Institute of Technology in Tamil Nadu created a model using artificial intelligence (AI) to translate the American Sign Language (ASL) into English instantly.

The Engineering student shared details about the project via her LinkedIn post, which reads, “Made this model using this cool Tensorflow object detection API. The model translates a few ASL signs to English using transfer learning from a pretrained ssd_mobilenet model.” Along with it, she shared a demo video that showcases the AI-based ASL Detector in action. And within no time, her post went viral on social media which garnered praise for her idea, which bridges the communication gap between the duo using sign language and English.

This is how an Indian student developed the AI model..

Though the AI model translates some basic sign codes into English such as Hello, I Love You, Thank you, Please, Yes, and No. In an interview with Interesting Engineering, she explained, "The dataset is made manually by running the Image Collection Python file that collects images from your webcam for or all the mentioned below signs in the American Sign Language: Hello, I Love You, Thank you, Please, Yes and No.

Gupta accepts that her AI model uses a transfer learning method which is trained by other experts. In the response to a comment on her LinkedIn post, she acknowledged "to build a deep learning model solely for sign detection is a really hard problem but not impossible and currently I'm just an amateur student but I am learning and I believe sooner or later our open source community which is much more experienced and learned than me will find a solution and maybe we can have deep learning models solely for sign languages."

She looks at it as a small baby step towards the bigger change. Her AI model is an idea to include and accept sign language on a smaller scale. Also, she dedicated the project to the data scientist Nicholas Renotte's tutorial video on real-time sign language.