AI Analyzes 119 Emergency Reports First... Fire Agency's Systematized Response and Support

ETRI Established and Demonstrated at Daejeon Gasuwon Safety Center

[Asia Economy Reporter Cha Min-young] Artificial intelligence (AI) technology analyzes 119 emergency call content to assist firefighters' response and support. Since 119 call reception involves a lot of noise due to the nature of the scene and most callers are in a tense state, it is expected to help field personnel quickly grasp the call content and disaster situation.

The Electronics and Telecommunications Research Institute (ETRI) announced on the 8th that it has established an AI-based 119 call reception system at the Gasuwon Safety Center of the Daejeon Fire Headquarters and has begun demonstration.

The system developed by the research team analyzes 119 call big data and applies AI technology to provide functions such as conversational speech recognition, situation-specific question recommendations for operators, disaster classification, and automatic response information provision. It displays the caller's voice content as text, shows disaster occurrence location information, and provides information such as automatic disaster classification according to the Emergency Rescue Standard Disaster Classification System.

It is expected to contribute to securing the golden time by shortening the emergency call reception processing time. ETRI received 130,000 pseudonymized 119 call reception voice data and 16,000 control history data from the demonstration institution, Daejeon Fire Headquarters, and conducted preprocessing and data analysis.

A preliminary satisfaction survey was conducted with 50 personnel involved in 119 call reception work regarding the research team's technology, and 85% responded positively that it would be helpful, expressing a strong hope for the rapid establishment of the system.

Going forward, additional research will be conducted on speech recognition text conversion technology specialized for real-time 119 call environments, deep learning-based 119 call reception disaster situation recognition and response support modeling technology, and AI-based 119 call reception and dispatch command support systems, with plans to develop a system for use by the demand agency, the National Fire Agency, by 2023.

Jung Eui-seok, head of ETRI's Public Safety Intelligence Research Lab, stated, "We plan to apply this technology to telephone and chatbot response tasks for the protection of citizens' lives and safety, significantly enhancing the public safety service satisfaction experienced by the people."