International Journal of Scientific Research and Engineering Development

Domestic violence is a serious social issue affecting millions of individuals worldwide, leading to physical, emo-tional, and psychological harm. Traditional reporting and monitoring methods are often delayed, inefficient, and lack real-time intervention mechanisms. This paper presents the design, implementation, and evaluation of a Do-mestic Violence Detection System—an intelligent, AI-based monitoring platform developed to identify violent activities through audio analysis, emotion recognition, and emergency alert mechanisms. The system is designed to assist victims by continuously monitoring voice patterns such as screaming, crying, aggressive speech, and abnormal sounds using machine learning and deep learning techniques. The proposed system supports real-time detection, automatic SOS alert generation, location sharing, and incident logging for rapid emergency response. The system is developed using Python, TensorFlow, OpenCV, Flask, and MySQL for efficient data processing and management. Experimental evaluation demonstrates significant improvements compared to traditional reporting methods: detection response time reduced from 20–30 minutes to under 1 minute, alert accuracy increased to 94%, false detection rate reduced below 5%, and emergency response efficiency improved substantially. The proposed system establishes a scalable, intelligent, and reliable framework for modern domestic violence prevention and victim protection

Prof. Chetan Patil, Janhavi Patil, Mrunal Jadhav, Yash Potdukhe, Kartiki Gaikwad,"Domestic Violence Detection System" International Journal of Scientific Research and Engineering Development, V9(3): Page(2523-2537) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.