Deep neural networks (DNNs) have gained remarkable success in speech recognition, partially attributed to the flexibility of DNN models in learning complex patterns of speech signals. This flexibility, however, may lead to serious over-fitting and hence miserable performance degradation in adverse acoustic conditions such as those with high ambient noises. We propose a noisy training approach to tackle this problem: by injecting moderate noises into the training data intentionally and randomly, more generalizable DNN models can be learned. This 'noise injection' technique, although known to the neural computation community already, has not been studied with DNNs which involve a highly complex objective function. The experiments presented in this paper confirm that the noisy training approach works well for the DNN model and can provide substantial performance improvement for DNN-based speech recognition.
A fault diagnosis method of wind turbine bearing based on intrinsic time-scale decomposition (ITD) is put forward. In the proposed method, the vibration signal of the main bearing is decomposed into several proper rotation components by the ITD method. The frequency centers of the proper rotation components that contain predominant energy are computed and considered as fault feature vectors. The nearest neighbor algorithm is applied to identify the fault types of the wind turbine bearing. The experimental data of the wind turbine spherical roller bearing in four conditions (normal, outer race fault, inner race fault and roller fault) are applied to evaluate the performance of the proposed method. The results demonstrate the feasibility and accuracy of this approach for the diagnosis of the wind turbine bearing faults under uncertain conditions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.