An Efficient Algorithm for Pitch Determination of Speech Signals - Kalman Filter Approach

In this paper, an efficient algorithm for pitch determination of speech signals is presented. Pitch period of speech signals does not contain sharp changes in time in voiced and voiced-unvoiced transition regions. Depending on this property, Kalman Filter has been used to estimate the pitch period similar to the way it is used in the target tracking problems. The measurement for the Kalman Filter is obtained by using the integer pitch calculation based on autocorrelation method, which is used as the first step of pitch determination in MELP speech coding algorithm. Kalman Filter makes an a priori estimate, which depends on the general behavior of the pitch period in time and this estimate is updated using this measurement to obtain an a posteriori estimate. The proposed method provides a reduction in the computational complexity since the autocorralation search is made only inside the gating volume of the Kalman Filter. Therefore, the proposed method does not require any pitch doubling check. It also does not need any fractional pitch computations. The pitch periods obtained using this method have been compared to those determined in MELP algorithm and it has been observed that comparible results have been obtained.