To perform a sound conversation processing based on processing data created by a display language.

In the sound conversation processing, a sound conversation processing part reads processing data created by the display language and performs a processing for emitting "Please input a member number" in accordance with a character filled after a "form" tag. Then, a system becomes a waiting state of sound input from a user in accordance with an "input" tag. When sound data is inputted, a sound recognition part performs a sound recognition processing and acquires character data. The sound conversation processing part holds character data received from the sound recognition part 12 in "members_number". Thus, the sound conversation processing is performed based on processing data created by the display language. Consequently, processing data for performing the sound conversation processing can be created by using the display language which is widely used.