Abstract
In this work a tele-operated system for a robot arm is designed and controlled by the operator, in witch voice control mode occur to carry out remote tasks. A voice command system for the manipulator arm is implemented as a high level control design and compared to a low level control design based on joystick. The high level control paradigm adopted is based on a spotted words recognition system using stochastic model, robust HMM (Hidden Markov Model) with Cepstral coefficients as parameters witch is used in automatic speech recognition system. To implement the approach on a real-time application, a Personal Computer USB interface was designed to transmit commands via Bluetooth components to control the movement of a Robot Arm (the TR-45). The user can control the movements of four degree of freedom (DOF) for TR-45 using a vocal phrase containing spotted words or using a joystick.