On the applications of the interacting multiple model algorithm for enhancing noisy speech

摘要

The interacting multiple model (IMM) algorithm is applied to enhancing speech contaminated by additive white or colored noise. Noisy speech is modeled by a linear state-space model with Markovian switching parameters. The parameters are estimated from the training speech and noise processes. The simulation results shows that the proposed method offers performance gains relative to the previous results with slightly increased complexity.