The game apparatus stores music data which indicates at least a pitch of each of sounds which form a melody of a predetermined piece of music and an output timing to output each of the sounds. The game apparatus sequentially detects, among the sounds included in the music data, a target sound which is a sound an output timing of which comes after start of playing by the music data. In the case where a first input is performed when or after the output timing of the target sound comes, the game apparatus outputs the target sound. On the other hand, in the case where a second or later input is performed after the output timing of the target sound comes, the game apparatus determines a pitch of an ad-lib sound, and outputs the ad-lib sound at the determined pitch.