Robust speech recognition system using bidirectional Kalman filter. Issue 6 (1st August 2015)

• Record Type: Journal Article
• Authors: Goh, Yeh Huann; Raveendran, Paramesran; Goh, Yann Ling
• Is Part Of: IET signal processing. Volume 9:Issue 6(2015)
• Page Start: 491
• Page End: 497
• Publication Date: 2015-08-01
• Subjects: speech recognition -- Kalman filters -- speech enhancement -- autoregressive processes -- state estimation -- mean square error methods -- difference equations -- matrix algebra -- hidden Markov models; robust speech recognition system -- bidirectional Kalman filter -- speech quality enhancement -- speech signal -- autoregressive process -- state‐space domain -- AR coefficient -- speech processing system -- system dynamics model -- state estimation -- mean squared error minimisation -- difference equation -- measurement equation -- constant matrix -- filtering process -- hidden Markov model; Signal processing -- Periodicals; 621.3822
• DOI: 10.1049/iet-spr.2014.0109
• Languages: English
• ISSNs: 1751-9675

Abstract

Abstract : Kalman filter is normally used to enhance speech quality in a noisy environment, in which the speech signals are usually modelled as autoregressive (AR) process, and represented in the state‐space domain. It is a known fact that to identify the changing AR coefficients in every time state requires extensive computation. In this paper, the authors develop a bidirectional Kalman filter and apply it in a speech processing system. The proposed filter uses a system dynamics model that utilises the past and the future measurements to form an estimate of the system's current time state. It provides efficient recursive means to estimate the state of a process that minimises the mean of the squared error. Compared to the conventional Kalman filter, the proposed filter reduces the computation time in two ways: (i) by avoiding the computation of AR parameters in each time state, and (ii) by reducing the dimension of the matrices involved in the difference equations and the measurement equations into constant (1 × 1) matrices. The speech recognition result shows that the developed speech recognition system becomes more robust after the proposed filtering process, and the proposed filter's low computational expense makes it applicable in the practical hidden Markov model‐based speech recognition system.