Over the past decades, extensive research has been carried out on various possible implementations of automatic speech recognition (ASR) systems. The most renowned algorithms in the field of ASR are the mel-frequency cepstral coefficients and the hidden Markov models. However, there are also other methods, such as wavelet-based transforms, artificial neural networks and support vector machines, which are becoming more popular. This review article presents a comparative study on different approaches that were proposed for the task of ASR, and which are widely used nowadays. † training time increases linearly with increase in vocabulary size [42] † quantisation error in the discrete representation of speech signals [42] † temporal information is ignored [42] PCA † reduction in the feature vector's size, while retaining much of the significant information [131] † robust [59, 60] † computationally expensive for high-dimensional data [8] LDA † maximises the distance between classes, but minimises the within class distance [132] † robust [133] † sample distribution is assumed a priori to be Gaussian [63] † class samples are assumed to have equal variance [63] Classification technique Advantages Disadvantages HMM † able to model time distribution of speech signals [103] † simple to adapt [68] † capable to model a sequence of discrete or continuous symbols [13] † inputs can be of variable length [40] † based on the assumption that the probability of being in a particular state is dependent only on its preceding state, ignoring any long-term dependencies [82] † emission probabilities are arbitrarily chosen; hence, these might not even represent properly the output probabilities of the corresponding state [82] ANN (in general) † good classifiers [16, 45] † highly adequate for pattern recognition applications [16, 45] † self-organising [16, 45] † self-learning [16, 45] † self-adaptive in new environments [16, 45] † robust [7] † based on ERM; hence, prone to over training a local minima problems [45, 103] MLP † good discriminating ability [2] † unable to model time distribution of speech signals [2] † inputs have to be of fixed length [2] † able to deal with small vocabularies only [2] SOM † no a priori information is required for training a SOM [134] † can easily adapt if a new sample is presented to it [134] † capable of parallel computation [134] † SOM algorithm is not well defined mathematically; hence, values for the network parameters need to be found by trial-and-error [134] † ordered mapping obtained after the training phase may be lost when applied in real environments due to frequent adaptations [134] RBF † simple to implement [135] † Good discriminating ability [135] † robust [135] † online learning ability [135] † shift invariant in time [91] RNN † able to model time distribution of speech signals thanks to the feedback connections [95, 103] † complex training algorithm [94] † training algorithm is highly sensitive to any changes [94] FNN † does not need large amount of samples during the learning process [99] † ...
Microgrippers are typical microelectromechanical systems (MEMS) that are widely used for micromanipulation and microassembly in both biological and micromanufacturing fields. This paper presents the design, modelling, fabrication and experimental testing of an electrothermal microgripper based on a ‘hot and cold arm’ actuator design that is suitable for the deformability characterisation of human red blood cells (RBCs). The analysis of the mechanical properties of human RBCs is of great interest in the field of medicine as pathological alterations in the deformability characteristics of RBCs have been linked to a number of diseases. The study of the microgripper’s steady-state performance is initially carried out by the development of a lumped analytical model, followed by a numerical model established in CoventorWare® (Coventor, Inc., Cary, NC, USA) using multiphysics finite element analysis. Both analytical and numerical models are based on an electothermomechanical analysis, and take into account the internal heat generation due to the applied potential, as well as conduction heat losses through both the anchor pads and the air gap to the substrate. The models are used to investigate key factors of the actuator’s performance including temperature distribution, deflection and stresses based on an elastic analysis of structures. Results show that analytical and numerical values for temperature and deflection are in good agreement. The analytical and computational models are then validated experimentally using a polysilicon microgripper fabricated by the standard surface micromachining process, PolyMUMPs™ (Durham, NC, USA). The microgripper’s actuation is characterised at atmospheric pressure by optical microscopy studies. Experimental results for the deflection of the microgripper arm tips are found to be in good agreement with the analytical and numerical results, with process-induced variations and the non-linear temperature dependence of the material properties accounting for the slight discrepancies observed. The microgripper is shown to actuate to a maximum opening displacement of 9 μm at an applied voltage of 3 V, thus being in line with the design requirement of an approximate opening of 8 μm for securing and characterising a RBC.
This review paper presents a comparative study of published integrated submicron CMOS quadrature voltage-controlled oscillator designs, based on LC resonator tanks operating at gigahertz frequencies. Although special reference to phase noise reduction is made, the comparison also concerns issues such as power consumption, tuning range and the phase accuracy of the quadrature signals. The effect of supply voltage reduction on the choice of the oscillator topology is also included in the discussion.
Microelectromechanical systems (MEMS) are the instruments of choice for high-precision manipulation and sensing processes at the microscale. They are, therefore, a subject of interest in many leading industrial and academic research sectors owing to their superior potential in applications requiring extreme precision, as well as in their use as a scalable device. Certain applications tend to require a MEMS device to function with low operational temperatures, as well as within fully immersed conditions in various media and with different flow parameters. This study made use of a V-shaped electrothermal actuator to demonstrate a novel, state-of-the-art numerical methodology with a two-way coupled analysis. This methodology included the effects of fluid–structure interaction between the MEMS device and its surrounding fluid and may be used by MEMS design engineers and analysts at the design stages of their devices for a more robust product. Throughout this study, a thermal–electric finite element model was strongly coupled to a finite volume model to incorporate the spatially varying cooling effects of the surrounding fluid (still air) onto the V-shaped electrothermal device during steady-state operation. The methodology was compared to already established and accepted analysis methods for MEMS electrothermal actuators in still air. The maximum device temperatures for input voltages ranging from 0 V to 10 V were assessed. During the postprocessing routine of the two-way electrothermal actuator coupled analysis, a spatially-varying heat transfer coefficient was evident, the magnitude of which was orders of magnitude larger than what is typically applied to macro-objects operating in similar environmental conditions. The latter phenomenon was correlated with similar findings in the literature.
A complete analysis of the spur characteristics of edge-combining delay-locked loop (DLL)-based frequency multipliers is presented in this brief. The novelty of this analysis is the fact that it can be used to estimate the effect of both the in-lock error and the delay-stage mismatch on the spurious level of the frequency multiplier with low computational complexity. In addition, a way to reduce the mismatch between the delay cells in the delay line is discussed via an analytic model and verified by the implementation of a delay cell in a 65-nm CMOS process.Index Terms-Delay mismatch, delay-locked loop (DLL), frequency synthesis, in-lock error, spurious level.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.