An adjustable active noise control algorithm is proposed based on human auditory perception. Because of the FELMS algorithm, the Bark scale decomposition is used for simulating the decomposition of the human ears to the acoustic signal, and the real time response of the human ears to the noise is calculated. Residual filter is designed based on this real-time response so that the noise reduction effect is more in line with the human beings auditory perception. In the meantime, the adjustment parameter is set to adjust the amplitude-frequency response of the designed residual filter so that the noise reduction effect can be adjusted artificially under the condition of preserving the noise characteristics. Computer simulation results show that, match with the convention FXLMS algorithm, the algorithm put forward in this paper makes the residual noise, and the original noise has a similar energy change trend in Barks. This algorithm makes the noise reduction effect more aligned with the human auditory perception. It can retain the noise characteristics and artificially adjust the noise reduction effect.