Communicating with someone who cannot speak or hear can be challenging and even though ASL (American Sign Language) exists, only a few learn to communicate using it.
Also Read:?5 Amazing Gadgets That Are Helping People With Hearing Impairment
However, browsing through LinkedIn, I stumbled upon a post by one Priyanjali Gupta that had developed an AI model that translated some ASL signs to English, bridging the gap between the divide.
Gupta, who is currently a student at the prestigious Vellore Institute of Technology, has been able to achieve this by harnessing Tensorflow object detection api. At its heart, it¡¯s making use of transfer-learning via a pre-trained model dubbed ssd_mobilenet.
In her Github post, she has revealed that the dataset was made manually by running a file (dubbed Image Collection python file from her GitHub repository) that collects images of ASL actions via a webcam.
As of now, it is capable of recognising six gestures -- Hello, I Love You, Thank You, Please, Yes, No.
In a comment to one of the LinkedIn users, she has acknowledged to the fact that creating a deep learning model from scratch just for sign detection is quite complex, ¡°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.¡±
Also Read:?Seven-Year-Old Mumbai Girl Built An App To Help Deaf Kids Easily Learn Sign Language
She added, ¡°Therefore for a small dataset and a small scale individual project I think object detection did just fine. It's just the idea of inclusion in our diverse society which can be implemented on a small scale.¡±
While this is still at a nascent stage, it¡¯s commendable that developers are looking towards creating more inclusive applications to help those in need.?
Keep reading?Indiatimes.com?for the latest?science and technology?news.