IJISA Vol. 4, No. 8, Jul. 2012
Cover page and Table of Contents: PDF (size: 141KB)
In speech synthesis accurate modeling of prosody is important for producing high quality synthetic speech. One of the main aspects of prosody is phone duration. Robust phone duration modeling is a prerequisite for synthesizing emotional speech with natural sounding. In this work ten phone duration models are evaluated. These models belong to well known and widely used categories of algorithms, such as the decision trees, linear regression, lazy-learning algorithms and meta-learning algorithms. Furthermore, we investigate the effectiveness of Support Vector Regression (SVR) in phone duration modeling in the context of emotional speech. The evaluation of the eleven models is performed on a Modern Greek emotional speech database which consists of four categories of emotional speech (anger, fear, joy, sadness) plus neutral speech. The experimental results demonstrated that the SVR-based modeling outperforms the other ten models across all the four emotion categories. Specifically, the SVR model achieved an average relative reduction of 8% in terms of root mean square error (RMSE) throughout all emotional categories.[...] Read more.
Rough set theory was introduced by Pawlak as a model to capture impreciseness in data and since then it has been established to be a very efficient tool for this purpose. The definition of basic rough sets depends upon a single equivalence relation defined on the universe or several equivalence relations taken one each at a time. There have been several extensions to the basic rough sets introduced since then in the literature. From the granular computing point of view, research in classical rough set theory is done by taking a single granulation. It has been extended to multigranular rough set (MGRS) model, where the set approximations are defined by taking multiple equivalence relations on the universe simultaneously. Multigranular rough sets are of two types; namely optimistic MGRS and pessimistic MGRS. Topological properties of rough sets introduced by Pawlak in terms of their types were studied by Tripathy and Mitra to find the types of the union, intersection and complement of such sets. Tripathy and Raghavan have extended the topological properties of basic single granular rough sets to the optimistic MGRS context. Incomplete information systems take care of missing values for items in data tables. MGRS has also been extended to such type of incomplete information systems. In this paper we have carried out the study of topological properties of pessimistic MGRS by finding out the types of the union, intersection and complement of such sets. Also, we have provided proofs and examples to illustrate that the multiple entries in the table can actually occur in practice. Our results hold for both complete and incomplete information systems. The multiple entries in the tables occur due to impreciseness and ambiguity in the information. This is very common in many of the real life situations and needed to be addressed to handle such situations in efficient manner.[...] Read more.
“Rule number explosion” in fuzzy controller and “uncertainty” in the model are two main issues in the design of fuzzy control systems. To overcome these problems, we have applied a method in which a linear sensory fusion function has been used to reduce the number of dimensions of fuzzy controller’s inputs and simultaneously use the features of LQR control. Since, in type-2 fuzzy control, the degree of fuzziness increased and it can better handle the uncertainty in the model compared to conventional fuzzy, so the method of sensory fusion with type-2 fuzzy control scheme has been combined to make the controller more robust w.r.t. the parameter variation, perturbance and uncertainty in the model. Performance criteria like IAE, ISE and ITAE have been used to compare the control performance obtained from conventional fuzzy and type-2 fuzzy controller.[...] Read more.
Human Face Recognition systems are an identification procedure in which a person is verified based on human traits. This paper describes a fast face detection algorithm with accurate result. Lip Tracking is one of the biometric systems based on which a genuine system can be developed. Since the uttering characteristics of an individual are unique and difficult to imitate, lip tracking holds an advantage of making the system secure. We use pre- recorded visual utterance of speakers has been generated and stored in the database for future verification. The entire project occurs in four different stages in which the first stage includes obtaining face region from the original image, the second stage includes mouth region extraction by background subtraction, the third stage includes key points extraction by considering the lip centroid as origin of co-ordinates and the fourth stage includes storing the obtained feature vector in the database. The user who wants to be identified by the system provides the new live information, which is then compared with the existing template in the database. The feedback provided by the system will be ‘a match or a miss-match’. This project will increase the accuracy level of biometric systems.[...] Read more.
In recent times machine learning algorithms are used for internet traffic classification. The infinite number of websites in the internet world can be classified into different categories in different ways. In educational institutions, these websites can be classified into two categories, educational websites and non-educational websites. Educational websites are used to acquire knowledge, to explore educational topics while the non-educational websites are used for entertainment and to keep in touch with people. In case of blocking these non-educational websites students use proxy websites to unblock them. Therefore, in educational institutes for the optimum use of network resources the use of non-educational and proxy websites should be banned. In this paper, we use five ML classifiers Naïve Bayes, RBF, C4.5, MLP and Bayes Net to classify the educational and non-educational websites. Results show that Bayes Net gives best performance in both full feature and reduced feature data sets for intended classification of internet traffic in terms of classification accuracy, recall and precision values as compared to other classifiers.[...] Read more.
Tyre building production line gradually transits from stand-alone production to combined production mode. The transformation of work mode from traditional serial intermittent to the parallel continuous has been the key technology research of tire production enterprise. And intelligent robots and other automated equipment have been the first choice of the tyre enterprise’s production line. Considering the combination of the equipment between upper and lower processes in tyre production line, the manual operations in some processes replaced by intelligent robots will improve production efficiency of tyre production enterprise, and will make outstanding contribution in reduce process losses and reduce production costs. This article studies on the key technology of combined application in production line, and makes simulation comparison for the same technical process that using different scheme, to prove the priorities and superiorities of combined production line relatives to the traditional production mode.[...] Read more.
Generalized Adaptive Linear Element (GADALINE) Artificial Neural Network (ANN) as an Artificial Intelligence (AI) technique is used in this paper to online adaptive control of a Non-linear Inverted Pendulum (IP) system. The ANN controller is designed with specifications as: network type is three (Input, Hidden and Output) layered Feed-Forward Network (FFN), training is done by Widrow-Hoffs delta rule or Least Mean Square algorithm (LMS), that updates weight and bias states to minimize the error function. The research is focused on how to adapt the control actions to solve the problem of “parameter variations”. The method is applied to the Nonlinear IP model with the application of some uncertainties, and the experimental results show that the system responds very well to handle those uncertainties.[...] Read more.
This paper has the following contributions in iris recognition compass: first, novel parameters selection for Gabor filters to extract the iris features. Second, due to iris textures randomness and assigning the Gabor parameters by pre-knowledgeable values, traditionally, a large Gabor filter bank has been used to prevent losing the discriminative information. It leads to perform extracting and matching the features heavily and on the other hand, the generated feature vectors are lengthened as required for extra storage space. We have proposed and compared two different approaches based on Genetic Algorithm to reduce the system complexity: optimizing the Gabor parameters and feature selection. Third, proposing a novel encoding strategy based on the texture variations to generate compact iris codes. The experimental results show that generated iris codes by optimizing the Gabor parameters approach is more distinctive and compact than ones based on feature selection approach.[...] Read more.