Amman — In a world of accelerating digital transformation, artificial intelligence is no longer merely a technology designed to make everyday life easier. It has become a promising and effective tool for rethinking inclusive education and developing ways that enable every learner to learn, communicate and express themselves according to their abilities, capacities and needs.
اضافة اعلان
One area that could undergo profound change thanks to these technologies is hearing disability, where many students face challenges in communicating at school, accessing information and interacting directly with teachers and peers.
Against this backdrop, the idea of developing an AI-powered assistant capable of analysing and understanding sign language within its educational context is gaining importance. Such an assistant would not simply recognise hand movements and convert them into written or spoken words. Rather, it would seek to understand the meaning conveyed by a sign in a classroom situation, taking into account facial expressions, body movements, the sequence of signs and the relationships between concepts. This represents an attempt to move from technology that sees movement to technology that helps understand the message.
Why do we need an intelligent assistant that understands sign language?
Sign language is a natural language with its own linguistic systems, rules and cultural characteristics. It is not merely a collection of movements corresponding to spoken words. A single movement can carry different meanings depending on the context, while facial expressions and movements of the head and body play an important role in conveying grammatical and emotional meanings.
In educational settings, deaf students may face difficulties when a teacher is not proficient in sign language or when learning materials are provided in formats that do not accommodate their linguistic needs. This gap may limit classroom participation, delay access to information or leave students dependent on a human intermediary in various situations.
The problem, however, does not lie with deaf students or sign language itself, but with the failure of some educational environments to accommodate linguistic diversity and different forms of communication.
This is where an intelligent assistant becomes important as a supportive tool that could help reduce communication barriers and create greater opportunities for interaction between students and teachers, without replacing a sign language interpreter.
How would the proposed AI assistant work?
The idea involves developing an intelligent system that can receive signs through a computer camera or tablet, then analyse visual information using computer vision, deep learning and natural language processing. The system could operate through several interconnected stages:
Recognising signs and movements: The system analyses the movements of the hands and fingers, their directions and their positions.
Analysing facial expressions and body movements: Understanding would not be limited to the hands, as facial expressions, head movements and body posture may help determine the type or meaning of a sentence, as well as some of its linguistic elements.
Understanding the educational context: The system uses the lesson topic, vocabulary and educational situation to interpret sign language in context. A sign used in a science lesson, for example, may have a different meaning or require a different translation from a similar sign used in an Arabic or mathematics lesson.
Challenges in developing an assistant that understands sign language
Despite its promising potential, developing an accurate and reliable system presents significant scientific and educational challenges. These include the diversity of sign languages, as there is no single universal language used by deaf people, as well as the need to understand context rather than simply memorise movements. A system may successfully recognise an individual sign but make a mistake when the speed of the movement changes.
By Dr. Ibrahim Bani Hamdan: A special education specialist.
