Priyanjali Guptaa third-year computer science student with a specialization in data science at Vellore Institute of Technology, has made notable progress in the field ofartificial intelligencewith a focus on inclusivity.
Its main innovation is an artificial intelligence model capable of translate American Sign Language (ASL) into English in real timedemonstrating how technology can help bridge the communication gap between people who are deaf or hard of hearing and those who don’t know sign language.
The inspiration for this project came from a challenge posed by Gupta’s mother, who urged her to use her engineering skills to create something concrete and useful. This inspiration led Priyanjali to think about how she could use her knowledge to help the deaf community.
He wants to train the model not only on single frames, but also on videos
After a year of intense work, he developed his own model based on deep learningusing the TensorFlow object detection API and applying transfer learning techniques through the pre-trained ssd_mobilenet model.
The model was trained to recognize six basic ASL signs: “Hello”, “I love you”, “Thank you”, “Please”, “Yes” and “No”. Using a simple webcams connected to the computer, the system recognizes these signs and translates immediately in English text, allowing for more fluid communication between those who use sign language and those who don’t know it.
Despite the widespread success of the project, which has garnered over 58,000 positive reactions on LinkedIn, Gupta is aware of the challenges of developing an artificial intelligence system complex like this one.
Moving forward, Gupta plans to improve the model by training it using Long-Short Term Memory (LSTM) networks to improve accuracy and enable recognition of sequences of moving signs. This approach could represent a major breakthrough for real-time sign language detection.
Priyanjali’s invention sparked great interest in the tech community and demonstrated what artificial intelligence can be used to promote inclusivity. If it is often maligned because it can take away jobs, in this case (and in many others) it has proven to be truly useful in breaking down barriers with a great positive impact on society.
