Speech recognition in a reverberant environment using matched filter array processing and linguistic-tree maximum likelihood linear regression adaptation

Abstract

Performance of automatic speech recognition systems trained on close-talking data suffers when the systems are used in a distant-talking environment due to the mismatch in training and testing conditions. Microphone array sound capture can remove some of the mismatch by removing ambient noise and reverberation, resulting in an approximation to a clean speech signal. However, this often does not improve the performance sufficiently. But, using array signal capture in conjunction with Hidden Markov Model (HMM) adaptation on the clean-speech models can result in high recognition accuracy. This paper describes an experiment in which the output of an eight-element microphone array system using MFA processing is used for speech recognition with LT-MLLR adaptation. The recognition is done in two passes. In the first pass, an HMM trained on clean data is used to recognize the speech. Using the results of this pass, the HMM model is adapted to the environment using the LT-MLLR algorithm. This adapted model is then used to recognize the speech. It is shown that the use of MFA and LT-MLLR results in high-accuracy recognition. [Work supported by DARPA Contract DABT63-93-C-0037.] a)Currently at Raytheon Systems Company, Falls Church, VA.

An application of speech/speaker recognition system for human-robot interaction

“…To overcome the reverberation of the sound, dereverberation technique and adaptation scheme were investigated. [1][2] Because those two are based on the inverse filter technique [1] and the adaptive filter design technique [2], previous researches are not appropriate to apply real time, recognition system.…”