A nonlinear adaptive noise canceller for speech enhancement using Volterra filter

Speech enhancement in the presence of very harsh noise is of practical importance, as conventional techniques and algorithms for speech denoising may suffer from serious performance degradation. A lot of real-life applications are calling for more sophisticated and powerful noise processing techniques and systems. Recently, both bone- and air-conducted speech signals have been used simultaneously to fight up against the background noise to improve the enhancement quality. Those enhancement systems consist of an adaptive noise canceller (ANC) that takes the bone-conducted speech as its reference signal while adopts the air-conducted measurement as its primary noise. A linear or nonlinear filter is implemented in the ANC. In this paper, we propose a nonlinear ANC by using of Volterra filter. The proposed system, enjoying nice convergence and computational efficiency, is capable of recovering the high-frequency components of the speech signal even in a very noisy setting. Simulations using real bone- and air-conducted speech measurements are performed to demonstrate the effectiveness of the proposed system.