Speech Augmentation using online Algorithms

Speech Augmentation aims to improve speech quality by using various techniques and algorithms . Minimising the intervening noise in a speech signal is a challenging task for many years. Compared to traditional unsupervised speech enhancement methods, e.g., Wiener filtering supervised approaches, such as algorithms based on hidden Markov models (HMM), lead to higher-quality enhanced speech signals. However, the main practical difficulty of these approaches is that for each noise type a model is required to be trained a priori. In this paper, we investigate a new class of supervised speech de noising algorithms using nonnegative matrix factorization (NMF). We propose a novel speech enhancement method that is based on a Bayesian formulation of NMF BNMF). To circumvent the mismatch problem between the training and testing stages, we propose two solutions. Firstly, HMM is used in combination with BNMF (BNMF-HMM) to derive a minimum mean square error (MMSE) estimator or the speech signal with no information about the underlying noise type. Secondly,a scheme is suggested to learn the required noise BNMF model online, which is then used to develop an unsupervised speech enhancement system.Extensive experiments are carried out to investigate the performance of the proposed methods under different conditions.