Noise suppression algorithm using Kalman filter with reduced computational complexity

We propose a noise suppression algorithm using Kalman filter with colored driving source. The algorithm aims to achieve robust noise suppression with reduced computational complexity by modifying the canonical state space models in. The remarkable features of the proposed algorithm are that it can be realized by 3 multiplications and that it has the better performances compared with despite the reduction of computational complexity, using only the Kalman filter algorithm for the proposed canonical state space models with the colored driving source: (i) a vector state equation is composed of the only speech signal, and (ii) a scalar observation equation is composed of speech signal and additive noise. We have confirmation of validity of the proposed canonical state space models with the colored driving source, and also show the effectiveness through numerical results and subjective evaluation results.