An electronic apparatus and method for configuration of an audio reproduction system is provided. The electronic apparatus receives image of a listening environment and applies a machine learning model on the image to identify a plurality of objects, including a display device and a plurality of audio devices. The electronic apparatus determines contour information of each identified object. The electronic apparatus retrieves real-dimension information of each identified object. Based on the determined contour information and the retrieved real-dimension information, the electronic apparatus determines first distance information between a listening position and each identified object. The electronic apparatus receives an audio signal from each audio device and determines a second distance between each audio device and the listening position based on the received audio signal. The electronic apparatus determines an anomaly in connection of at least one audio device and generates connection information based on the determined anomaly.