Framework for Voice-controlled AI Teaching Assistant for Specially-abled

Recent advancements in technology have triggered a revolution in the education sector. One of the significant developments of this change is being granted access to vast online resources from various world class institutes. The motivation behind this paper is to be able to create a technological solution for providing equal opportunities to specially-abled online learners who may have physical disabilities such as visual impairment, loss of hands/weak hand motor skills, etc., and enable their participation for a more inclusive society. The paper uses a twofold approach. Firstly, it provides literature on various state-of-art assistive technologies for e-learning portals. Secondly, the paper proposes a learning framework as teaching assistant for specially-abled, which is artificial intelligence based and interactive voice enabled. The proposed framework has three modules completely controlled by voice commands for self-paced -reading, learning and auto-marking assessment. The assessment module generates fill-in-the-blanks and question-answers, which is a part of continuous comprehensive tests. As a case study, the proposed framework is prototyped on an e-learning platform DIKSHA, which hosts digital books from National Council of Educational Research and Training. Natural language processing toolkit based on Python libraries has been used. The results have been found encouraging. Extensions for future work are also discussed.