The robot and human communicate with each other through interfaces. There are a wide variety of modalities, ranging from manual – from old-fashion Teach Pendants (TP) with light displays and input buttons, to touch screens with graphical programming languages –, voice – which include a microphone for the human and a speaker for the robot –, kinesthetic – where the human moves the robot by hand to teach desired positions or trajectories –, to visual – the human shows the robot what needs to be done by doing the task in front of a vision system. Depending on the task, a combination of interface modalities may bring the largest benefit.