Model Adaptation Based On HMM Decomposition For Reverberant Speech Recognition

The performance of a speech recognizer is degraded drastically in reverberant environments. We proposed a novel algorithm which can model an observation signal by composition of HMMs of clean speech, noise and an acoustic transfer function[1]. However, how to estimate HMM parameters of the acoustic transfer function is a remaining serious problem. In our previous paper[1], we measured real impulse responses of training positions in an experiment room. It is inconvenient and unrealistic to measure impulse responses for every possible new experiment room. This paper presents a new method to estimate HMM parameters of the acoustic transfer function from some adaptation data by using an HMM decomposition algorithm which is an inverse process of the HMM composition. Its effectiveness is confirmed by a series of speaker dependent and independent word recognition experiments on simulated distant-talking speech data