To provide a duplication processing apparatus capable of ensuring the continuity of voice sound to be duplicated.

A plurality of images respectively belonging to a plurality of channels are captured in a capturing area of a hard disk 50 and continuous speech common to the plurality of channels is allocated dispersively to a plurality of images stored in the capturing area. A CPU 44 selects each of the plurality of images stored in the capturing area in a designation sequence and discriminates whether a channel to which the selected image belongs is a designation channel. When a discrimination result is negative, partial speech allocated to the selected image is saved to a save area of an SDRAM 34. When the discrimination result is positive on the other hand, the selected image is duplicated to a duplicate area of the SDRAM 34, and the partial speech allocated to the selected image and the partial speech stored in the save area are duplicated in the duplicate area.