LPCs enhancement in iterative Kalman filtering for speech enhancement using overlapped frames

In this work, we are concerned by a new iterative Kalman filtering scheme where a linear predictor model parameters are estimated from noisy speech. However, when only noise-corrupted speech is available, the enhancement performance of the Kalman filter is somewhat dependent on the accuracy of the LPC and excitation variance estimates. Nevertheless, linear prediction based speech (LPC) analysis is known to be sensitive to the presence of additive noise. To overcome this problem we present in this paper an analysis and application of the iterative Kalman filtering with overlapped frames. Our enhancement experiments use a NOIZEUS corpus where the proposed method achieves higher Perceptual Evaluation of Speech Quality (PESQ) score and better subjective tests than the iterative scheme of Gibson as well as other enhancement methods.