An embodiment of the present invention provides a speech recognition engine that utilizes portable voice profiles for converting recorded speech to text. The supply comprises a carrier strip having at least an upper and a lower surface as well as and two longitudinally extending edges arranged parallel to each other. The second output signal is subtracted from the first output signal to generate a transmitter output signal. A manual gear input unit is selected for selecting gears with a first actuating element 2 for selecting at least one gear and with at least one additional actuating element 3 for setting additional functions. The present invention provides an apparatus and method for facilitating the processing of optical signals in planar waveguides received from optical fibers.