Hybrid CNN-Based Hand Gesture and Voice Control for Accessible Media Playback

Abstract

Human-computer interaction has evolved with the advancement of machine learning techniques such as Convolutional Neural Networks (CNNs) and speech recognition. This paper presents a hybrid system that allows users to control media playback using hand gesture recognition via CNN’s and voice commands. The system leverages a webcam to capture hand gestures and a microphone for speech input, enabling actions such as play, pause, next, previous, volume up, and volume down. The proposed dual modality approach enhances user convenience and accessibility, particularly for physically challenged users. Experiments show high accuracy in both gesture and voice recognition modules confirming legacy of the system