Hi people, is there some way to avoid the excess of information at the output, and make appears only the spoken word?