In smart cities, ensuring road safety and optimizing transportation efficiency heavily relies on streamlined road condition monitoring. The application of Artificial Intelligence (AI) has notably enhanced the capability to detect road surfaces effectively. This study presents a novel approach to road condition monitoring in smart cities through the development of an acoustic data processing and analysis module. It focuses on four types of road conditions: smooth, slippery, grassy, and rough roads. To assess road conditions, a microphone integrated road surface detector unit is designed to collect audio signals, and an ultrasonic module is used to observe the road depth information. The whole hardware unit is installed in the wheel rim of the vehicles. The data collected from the road surfaces are then analyzed using machine learning algorithms, such as Multi-Layer Perceptron (MLP), Support Vector Machine (SVM), and Random Forest (RF). The results demonstrate the effectiveness of the proposed method in accurately identifying different road conditions. From these results, it was observed that the MLP provides better accuracy of 98.98% in assessing road conditions. The study provides valuable insights into the development of a more efficient and reliable road condition monitoring system for delivering secure transportation services in smart cities.