Published Paper Details:

Emergency dispatch systems are important to provide timely and effective response to serious situations. Conventional call evaluation techniques are based on human assessment, which is subjective and unreliable. Voice stress analysis (VSA) has been found to be a potential technique for analyzing the stress level of callers and enhancing decision-making in emergency dispatch centers. This study delves into the incorporation of AI-based voice stress analysis in emergency call systems using machine learning and speech processing algorithms to identify stress patterns in real time. Through examination of fluctuations in pitch, tone, and speech rhythm, the system is able to evaluate the level of urgency and emotional state of the caller and assist dispatchers in prioritizing responses more effectively. Experimental testing with real and simulated emergency call datasets proves VSA's success in enhancing response times, decreasing misclassification of emergency severity, and increasing situational awareness. The results prove the potential for AI-driven voice stress analysis in maximizing emergency response strategies while responding to accuracy, privacy, and ethical concerns challenges.