In humans, speech impairment hinders one s capacity to talk and communicate with others, and such a condition forces one to adopt sign language as a means of communication. Communication between persons who under- stand sign language and those who do not is equally difficult. The majority of individuals do not understand sign language. Visually impaired people are unable to converse with them. As a result, our purpose with Divyang Sabalikarana is to make communication between us all simpler. This has the potential to have a significant impact on the disability community in particular. In this study, a Convolutional Neural Network (CNN)-based approach for detecting hand contact is developed. On numerous metrics, such as action time, accuracy,sensitivity, specificity,good and poor prediction,like lihood,and square root description, the enhanced approach is assessed and compared between training and testing modes. The results reveal that CNN is a successful approach to extract various characteristics and segregate data,with ate staccuracy of 99%.