Proposed Integration for AICC-based Chatbot Improvement

ABSTRACT

Artificial intelligence is the foundation of the era of the Fourth Industrial Revolution and has been applied to variou sfields. Among them, there are many fields that use artificial intelligence-based chatbots, such as finance, medical care, etc. C hatbots are also used in call centers, where the turnover rate of call center counselors is traditionally 13 times higher than tha tof other industries. In addition, it is difficult to maintain stable counseling due to realistic difficulties such as time and invest ment costs even when professional counselors are continuously fostered in call centers. Therefore, practical measures are requir ed to stabilize the supply and demand of manpower in the call center industry and support continuous growth. In this study, we present a customized AICC operation service structure that can alleviate the difficulties of call center counselors, receive c ustomer requests, and provide answers in the most similar form to humans instead of counselors. Until now, chatbots functios were based on voice recognition ARS, TTS, and ERMS for a quick response to simple services and connection to counselors. In contrast, the contents presented in this study suggest a service structure that can respond to customers within 1.4 seconds a fter a phone call by combining STT, NLU, and TTS with an AICC-based counseling task automation environment. The establ ished system can secure work efficiency and strengthen expertise, and a plan is presented for a system structure that enables future customer communication in the non-contact era as COVID-19. © 2022, Ismail Saritas. All rights reserved.