In this invention, a voice recognition engine outputs to a controlling section a matching state of a voice input signal as an error code. Then, the controlling section determines the matching state based on the error code in the error determination section and outputs to a voice synthesizing engine guidance data according to the matching state based on a timing control by a guidance timing controlling section. According to such a configuration, this invention improves operatability by voice operation, while reducing a risk of erroneous recognition by maintaining a predetermined matching rate.
