Disclosed herein is a method and a video generator for generating video response to user queries. The video generator receives a visual image of a character of interest from the user and generates a frontal face of the visual image. Further, facial expressions of the character of interest are mapped with an audio/video sequence of one or more textual responses for generating a human like video response to the user queries. In an embodiment, the video generator detects gender of the character of interest, and modulates and matches voice of the video response based on the gender of the character of interest. The instant method can synthesize a video with the face of a character of interest to the user, thereby providing a wholesome communication experience to the user.