Being able to express our thoughts, feelings, and ideas to one another is essential for human survival and development. The family of hearing-impaired members often has difficulty communicating with them and need to interact with them using sign language. Their family still faces many challenges in learning sign language despite technological advance. If the person is bedridden and hearing impaired, communication with the caretaker becomes a challenging task. Visual interpretation of gestures and speech along with facial expressions can be useful in accomplishing human-like interaction. The concept of an avatar emerges as a digital representation of a user’s presence in a virtual reality space, capable of interacting with a computer-generated environment or other users. By using avatars, hearing-impaired individuals can better understand the tone and intent of the communication, allowing them to respond appropriately and enabling them to have an inclusive social life. The proposed system facilitates a generation of facial animation of emotions along with gestures on a 3D avatar input speech given by a normal person. Deep learning is used to estimate the emotion and gestures from the input speech. These resultant gestures and facial emotions are represented through animations to the hearing impaired in a 3D Avatar.