A video doorbell and an automatic setting method thereof are provided. The video doorbell includes a bridge rectifying unit, a switch unit, and a processing unit. The bridge rectifying unit is configured to output a DC voltage when receiving an AC power from a transformer. During a chime type detection mode, the switch unit is configured to switch a path of the AC power for generating a short-circuit current to trigger a chime. After the short-circuit current is kept for a detection time, the short-circuit current is terminated. During the chime type detection mode, when the short-circuit current is terminated, the processing unit detects a rising time of the DC voltage rising from a first level to a second level, and determine whether the chime is an analog chime or a digital chime according to the rising time.