Interacting with a computer can be made easy through voice commands and hand gestures with the gesture controlled virtual mouse. Direct physical interaction with the computer is not really necessary. Combining vocal instructions with hand movements that are both static and dynamic could enable digital management of all input and output procedures. Modern machine learning (ML) and computer vision (CV) techniques are used in this study to identify spoken commands and hand gestures without the requirement for further hardware. It utilizes MediaPipe, which is predicated on pybind11, to employ CNN-like models. It consists of two modules: one uses MediaPipe hand detection to operate directly on hands, and the other person wears gloves of any color.