A factory automation system 1 comprises programmable logic controllers 3, 5, 7, a microphone 15, and a remote server 10 connected via a network 9. The microphone 15 is connected to the programmable logic controller 3, and devices 13 are connected to and controlled by the programmable logic controllers 3, 5, 7. The programmable logic controller 3 encodes spoken words of a user 19 recorded by the microphone 15 into an electronic speech representation. The electronic speech representation is transmitted to the remote server 10. The remote server 10 determines a device command based on the electronic speech representation and transmits the device command to the programmable logic controller 3. The programmable logic controller 3 controls the devices 13 based on the received device command.