Effectiveness of discriminative training and feature transformation for reverberated and noisy speech

Abstract

Automatic speech recognition in the presence of non-stationary interference and reverberation remains a challenging problem. The 2nd Annual Speech Separation and Recognition Challenge introduces a new and difficult task with time-varying reverberation and non-stationary interference including natural background speech, home noises, or music. This paper establishes baselines using state-of-the-art ASR techniques such as discriminative training and various feature transformation on the middle-vocabulary sub-task of this challenge. In addition, we propose an augmented discriminative feature transformation that introduces arbitrary features to a discriminative feature transformation. We present experimental results showing that discriminative training of model parameters and feature transforms is highly effective for this task, and that the augmented feature transformation provides some preliminary benefits. The training code will be released as an advanced ASR baseline. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)This work may not be copied or reproduced in whole or in part for any commercial purpose. Permission to copy in whole or in part without payment of fee is granted for nonprofit educational and research purposes provided that all such whole or partial copies include the following: a notice that such copying is by permission of Mitsubishi Electric Research Laboratories, Inc.; an acknowledgment of the authors and individual contributions to the work; and all applicable portions of the copyright notice. Copying, reproduction, or republishing for any other purpose shall require a license with payment of fee to Mitsubishi Electric Research Laboratories, Inc. All rights reserved. ABSTRACT Automatic speech recognition in the presence of non-stationary interference and reverberation remains a challenging problem. The 2 nd 'CHiME' Speech Separation and Recognition Challenge introduces a new and difficult task with time-varying reverberation and non-stationary interference including natural background speech, home noises, or music. This paper establishes baselines using state-of-the-art ASR techniques such as discriminative training and various feature transformation on the middle-vocabulary sub-task of this challenge. In addition, we propose an augmented discriminative feature transformation that introduces arbitrary features to a discriminative feature transformation. We present experimental results showing that discriminative training of model parameters and feature transforms is highly effective for this task, and that the augmented feature transformation provides some preliminary benefits. The training code will be released as an advanced ASR baseline.

Deep recurrent de-noising auto-encoder and blind de-reverberation for reverberated speech recognition

“…Large improvements can be obtained even with a clean recognizer back-end; furthermore, in unseen acoustic conditions the data-based method achieves notable performance compared to the model-based method. Future work will concentrate on the integration of discriminative methods both in the ASR back-end training and in the DAE training, which have proven effective for reverberated speech [28]. Furthermore, for better integration with the ASR back-end, we will investigate improved cost functions in DAE training taking account parameters of the ASR back-end instead of just optimizing distances in the spectral domain.…”

A study on automatic speech recognition toolkits

“…Yuuki Tachioka et al [12] verified the effectiveness of discriminate training under clean, reverberant and noisy speech using MFCC features, MFCC+LDA+MLLT features and PLP features separately. The feature transformation are effective on the non-stationary interference and reverberation.…”

Reverberant speech recognition exploiting clarity index estimation

“…In [13], a model of the noise is estimated from observed data by considering the late reverberation as additive noise, and then the feature vector is enhanced by applying vector Taylor series. A feature transformation based on discriminative training criterion inspired on Maximum Mutual Information is suggested in [14]. Additional features related to the amount of diffuse noise in each frequency bin and frame are employed in [15] to improve deep neural network-based ASR accuracy in noisy and reverberant environments.…”