Welcome to kNOw SIGN!
This is a project completed by a group of students at University of Glasgow, which employs utilization of real-time and efficient programming to help mute and deaf by establishing two-way communication channel with people who have never studied sign language.
According to the study conducted by the World Health Organization (WHO) in 2015, the deaf and mute constitutes about 5% of the world’s population, which is approximately 70 million. But effective communication is a hindrance for this community in the real world. The WHO predicts that by 2050, nearly 2.5 billion or 1 in 4 people will have some degree of hearing loss. The report notes that the use of sign language and its interpretation can further improve access to communication and education for those with hearing loss. A serious impediment for this community is the limited availability of written material and sign language interpreters at health services. It is said that millions benefit from the advancements in this field, however their use is still far too limited.
Hence, our goal was to create a system that aids in effective communication and to be the voice of the unheard. The most popular sign language in use is the American Sign Language(ASL). We incorporated a system that executes real-time ASL recognition and reproduces it onto an LED display as text. It utilizes Pi camera (V2.1) for hand detection attached to Raspberry Pi 4B, which acts as the computing unit. We aim this product to be an inexpensive alternative to other services while simultaneously realizing the objective of using fast and efficient deep learning algorithms. Furthermore, the system has a compact and portable design considering the wearable comfort of the user.
Further development in this work would include incorporating other types of sign languages, using more efficient models, and integrating a text to speech design.
Video Demonstration
Checkout the kNOw SIGN demo!