Deep Learning-Based Audio Classification for Smart Home Security

Traditional smart home security systems often rely on static sensors, limiting their ability to detect real-time unseen threats that can only be detected by acoustic processing. This research focuses on developing an advanced security system that identifies dangerous sounds and alerts homeowners via SMS when a potential threat is detected. The system is based on Raspberry Pi and integrated with a mobile robot with obstacle avoidance capabilities, allowing it to move through the house, record sounds, and analyze them for hazards such as breaking glass, gunshots, or baby crying. By utilizing Artificial Intelligence (AI) and Machine Learning (ML) techniques, particularly Convolutional Neural Networks (CNN), this solution enhances home security by providing timely notifications and proactive threat detection. Additionally, this system benefits individuals who are deaf or hard of hearing by offering alternative alerts, such as visual signals or vibration-based notifications, ensuring inclusivity in home security.