Published Paper Details:

Listened fed people face important challenges in the urban environment, as it is an inability to detect important hearing signs such as vehicle horns, emergency sirens and other traffic -related sounds. Navigating in busy roads without awareness of going to vehicles leads to a serious security risk that limits their dynamics and freedom. To solve the problem, the project suggests an IoT-based Help Smartwatch designed to detect real-time vehicle sounds and provide intuitive alerts through light signals and vibrations.The system includes a sound detection sensor, an ESP32 microcontroller and a machine learning-based classification models, which accurately identify the sound of the vehicle when filtering the background noise. A hybrid CNN- SVM model is used to ensure accurate sound classification, and separates a variety of traffic sounds such as car horns, ambulance sirens and heavy vehicle engines. When you detect a possible danger, the smart watch activates LED indicators and vibrates with different intensity depending on the urgency of the detected sound.