A method of teaching speech reading comprising the steps of displaying a first moving image of the facial articulatory structure of at least one person; displaying a second moving image of alphanumeric indicia representing the words spoken by said at least one person; and controlling the display of the first and second moving images for varying the time interval between the display of the alphanumeric indicia and the display of the particular facial articulatory structures corresponding thereto. Apparatus for carrying out the method is also disclosed.