In this paper a hybrid tunable vibration absorber has been made of shape memory materials. This vibration absorber is added to a main system of a cantilevered beam. Adaptive sliding mode control is applied, along with an artificial neural network to modeling the system's inverse dynamics, to suppress the system vibration by adjusting the vibration absorber stiffness and damping ratios when the system is subject to a single frequency external excitation. Experimental results are to be compared to those obtained by skyhook controller. It is shown experimentally that when disturbance frequency is varied away from the designed natural frequency of the vibration absorber, adaptive sliding mode control can effectively detune the system so as to avoid resonant vibration, hence ensuring adaptive and robust vibration suppression performance of the system.