A method for providing a composite voice-screen automatic response system (ARS) service, according to one embodiment of the present invention, comprises: a step for receiving, by a server for a multimedia call, a call request message from user equipment; a step for determining whether the server for the multimedia call possesses ARS contents corresponding to a call number of the call request message; a step for transmitting, by the server for the multimedia call, a command for downloading the ARS contents to the user equipment when it is determined that the server for the multimedia call possesses the ARS contents; and a step for transmitting, by the server for the multimedia call, the call request message to the call number, wherein the user equipment can receive an ARS service synchronized with a voice and a screen without installing an application according to each enterprise for providing the ARS service.