We present a discriminant measure that can be used to determine the model complexity in a speech recognition system. In the speech recognition process, sub-phonetic classes are modelled as mixtures of Gaussians, and in this correspondence we present a new discriminant measure that uses the classification accuracy to determine in an objective fashion, the number of Gaussians required to best model the pdf of an allophone class. We compare the performance of this criterion with other criteria such as BIC, and show that BIC and the discriminative criterion lead to parsimonious models that provide the same word error rate performance as much larger baseline systems. However, this performance improvement depends on the size of the system, and there appears to be a crossover point beyond which both BIC and the discriminative criterion are worse than a much simpler criterion. The discriminative criterion also enables this crossover point to be controlled by means of a threshold that is used in the criterion, and can lead to a better tradeoff of complexity versus word error rate.
This paper introduces the basic concepts of harmonics in power system, and expounds the basic principles of harmonic detection algorithms such as Fourier transform, short wavelet transform and wavelet packet transform. The effectiveness of the proposed algorithm is verified by simulation, which shows that the improved wavelet packet algorithm can achieve the uniform division of the harmonic signal frequency band, and improve the detection accuracy of the harmonic signal. Wavelet packet transform could detect the fundamental and harmonic components of the original signal, which is most suitable for the detection of harmonic signals.
In the majority of the multivariable processes, analysis of process monitoring and fault diagnosis is usually based on the fundamental assumption that the monitored variables follow a Gaussian distribution. However, it is well known that many of the variables are mutually dependent in process systems. This paper proposes a new monitoring method based on independent component analysis (ICA) -sparse autoencoder. The independent information component can be extracted by ICA through higherorder statistics. Moreover, the inherent nonlinear characteristics in the residual model of ICA can be handled by a deep architecture constructed with sparse autoencoder. To overcome the problem of local minima in the optimization of sparse autoencoder, a restricted Boltzmann machine (RBM) is used to pre-train the net, and the parameters in the sparse autoencoder is updated by Limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS). Monitoring statistic is developed and its confidence limit is computed by kernel density estimation. A case study of the Tennessee Eastman (TE) benchmark process indicates that the proposed fault detection method is more efficient.
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.