Abstract: A method for a speech input is provided. The method includes steps of: (a) establishing a hierarchical command list having a plurality of classified commands and a plurality of predetermined commands in a first device, (b) issuing a first speech command to the first device from a third device of a second device so that the first device obtains a first classified command corresponding to the first speech command, (c) determining a first command set from the plurality of predetermined commands via the first device based on the first classified command, (d) controlling the first device by the third device so as to cyclically provide each command in the first command set, (e) issuing a second speech command according to the each command in the first command set, (f) recognizing the second speech command with the first device, and (g) performing an operation corresponding to the recognized second speech command.