From Theory to Practice: Implementing Artificial Intelligence in Sign Language Instruction in Cameroon

This study chronicles our efforts in exploring challenges faced in traditional sign language teaching methods and proposes AI- driven solutions that effectively address these issues. We place this study in the framework of Technology Acceptance Model (TAM). This experimental study was carried out with 40 undergraduate university students in Bambili – North West Region of Cameroon.An experimental research design was employed involvingreal-time training using Neural Network (DenseNet 201 Model) online tool. Observation sessions and 40 questionnaires were used as instruments for data collection. Findings indicate that: 1.Traditional teaching methodscharacterized: Passive learning, heavy reliance on the teacher as the sole source of knowledge, inadequate opportunities for practical applications, rote memorization and repetition of gestures hindered smooth teaching and learning.2.The use of AI- driven tools: Video demonstrations for sign recognition includeddatasets of 100 images,26 letters of manual alphabets, 30 numbers, 50 fingerspelledwords, 40 common phrases and 25 basic signs, capturing handshapes and gestures, angles, positions, and proximity enhanced students learning skills.By adopting these modern teaching methods with AI tools, teachers create a learning environment that promotes active engagement, practical application and effective communication skills in sign language. The study recommends that Governments, educational institutions, and organizations should promote sign language and challenge negative stereotypes by encouraging innovative technologies that foster the teaching and learning of sign language.