Nordin Abu Bakar

An evaluation of endpoint detection measures for malay speech recognition of an isolated words

This paper presents the endpoint detection approaches specifically for an isolated word uses Malay spoken speeches from Malaysian Parliamentary session. Currently, there are 34,466 vocabularies of utterances in the database collection and for the purpose of this study; the vocabulary is limited to 25 words which are most frequently spoken selected from ten speakers. Endpoint detection, which aims to distinguish the speech and non-speech segments of digital speech signal, is considered as one of the key preprocessing steps in speech recognition system. Proper estimation of the start and end of the speech (versus silence or background noise) avoids the waste of speech recognition evaluations on preceding or ensuing silence. In this study, the endpoint detection and speech segmentation task is achieved by using the three different algorithms, namely combination between Short-time Energy (STE) and Zero Crossing Rate (ZCR) measures, frame-based Teagers Energy (FTE), and Energy-Entropy feature (EEF). Three experiments were conducted separately to investigate the overall recognition rate obtained with a Discrete-Hidden Markov Model (DHMM) classifier approach on the testing data set that consists of 1250 utterances. The results show that EEF algorithm performs quite satisfactory and acceptable where average recognition rate is 80.76% if compared with other two algorithms. Each of the algorithms have the advantages and disadvantages and there are still misdetection of word boundaries for the words with weak fricative, plosive and nasal sounds and not robust enough to implement in Malaysian Parliamentary speech data. However, improvement is still possible to increase the performance of these algorithms.

Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech

This paper presents the endpoint detection approaches specifically for an isolated word uses Malay spoken speeches from Malaysian Parliamentary session. Currently, there are 7,995 vocabularies of utterances in the database collection and for the purpose of this study; the vocabulary is limited to ten words which are most frequently spoken selected from ten speakers. Endpoint detection, which aims to distinguish the speech and non-speech segments of digital speech signal, is considered as one of the key preprocessing steps in speech recognition system. Proper estimation of the start and end of the speech (versus silence or background noise) avoids the waste of speech recognition evaluations on preceding or ensuing silence. In this study, the endpoint detection and speech segmentation task is achieved by using the short-time energy (STE) and short-time zero crossing (STZC) measures and combination of both approaches. As a result, the Hidden Markov Model (HMM) recognizer derived the recognition accuracy rate of 91.4% for combination of both algorithms, if compared only 86.3% for STE and 82.1% for STZC rate alone. The experiments show that there are many problems arise where there are still misdetection of word boundaries for the words with weak fricative and nasal sounds. Other obstacles issues such as speaking styles or mood of speaking can also cause the recognition performance.

DEPENDABILITY ATTRIBUTES FOR INCREASED SECURITY IN COMPONENT-BASED SOFTWARE DEVELOPMENT

Existing software applications become increasingly distributed as their continuity and lifetimes are lengthened; consequently, the users’ dependence on these applications is increased. The security of these applications has become a primary concern in their design, construction and evolution. Thus, these applications give rise to major concerns on t he capability of the current development approach t o develop secure systems. Component-Based Software Development (CBSD) is a software engineering approach. CBSD has been successfully applied in many domains. However, the CBSD capability to develop secure software applications is lacking to date. This study is an extension of the previous study on the challenges of the security features in CBSD models. Therefore, this study proposes a solution to the lack of security in CBSD models by highlighting the attributes that must be embedded into the CBSD process. A thorough analysis of exist ing studies is conducted to investigate the related software security attributes. The outcome analysis is beneficial for industries, such as software development companies, as well as for academic inst itutions. The analysis also serves as a baseline reference for companies that develop component-based software.

Measuring the performance of isolated spoken Malay speech recognition using Multi-layer Neural Networks

This paper describes speech signal modeling techniques which are suited to high performance and robust isolated word recognition. In this study, a speech recognition system is presented, specifically an isolated spoken Malay word recognizer which uses spontaneous and formally speeches collected from Parliament of Malaysia. Currently the vocabulary is limited to 25 words that can be pronounced exactly as it written and controls the distribution of the vocalic segments. The speech segmentation task is achieved by adopted energy based parameter and zero crossing rate measure with modification to better locates the beginning and ending points of speech from the spoken words. The training and recognition processes are realized by using Multi-layer Perceptron (MLP) Neural Networks with two-layer network configurations that are trained with stochastic error back-propagation to adjust its weights and biases after presentation of every training data. The Mel-frequency Cepstral Coefficients (MFCCs) has been chosen as speech extraction approach from each segmented utterance as characteristic features for the word recognizer. Recognition results showed that the performance of the two-layer networks increased as the numbers of hidden neurons increased. The best network structures average classification rate is 84.731% with (150-25) configuration. Implementation results also showed that the conjugate gradient (CG) algorithm was more accurate and reliable than the Levenberg-Marquardt (LM) algorithm for the network complexities and data size considered in this study.

AWARENESS OF EMBEDDING SECURITY FEATURES INTO COMPONENT-BASED SOFTWARE DEVELOPMENT MODEL: A SURVEY

Current applications and systems contain the software components as the basic elements and Component Based Software Development (CBSD) has been successful in building applications and systems. However, the security of CBSD for the software component is still lacking. This study highlights the results of a survey pertaining to the embedding of security features in the CBSD process. The main objective of this survey is to investigate the awareness of embedding security features in the CBSD process in the Malaysian context. For this purpose, experts from industry as well as from the academic community were interviewed. Moreover, an online survey was formulated and e-mailed to the experts and potential candidates. The results show that the embedding of security features in the software lifecycle is crucial because the incorporation of security activities in CBSD will minimize vulnerabilities in the software system, thus reducing system cost.

The optimization of Artificial Neural Networks connection weights using genetic algorithms for isolated spoken Malay parliamentary speeches

This paper presents the structure of a neural network models for validation recognition performances of isolated spoken Malay utterances. Artificial Neural Network (ANN) has been well recognized for its approximation capability provided the input-output data are available. Nevertheless, the conventional training algorithm, Levenberg-Marquardt (LM) algorithms that utilized as gradient search method in the model development has always encountered difficulties to converge at global solution. Aiming at improving the accuracy and robustness of ANN model, Genetic Algorithm (GA) was introduced in ANN modelling for connection weights evolution. From the results, it was observed that the performance of GA-ANN models is better than ANN-LM models. Integrating the GA with feedforward network can improve mean square error (MSE) performance and by this two stage training scheme, the recognition rate can be increased up to 85%.

Self-Regulated Learning and Online Learning: A Systematic Review

Self-regulated learning (SRL) is an academically effective form of learning, which learners must set their goals and make plans before starting to learn. As an ongoing process, learners need to monitor and regulate their cognition, motivation, and behavior as well as reflect on their learning process. These processes will be repeated as a cyclic process. The emerging technologies have changed the learning environments. Technology delivers teaching to learners via online. In online learning, information of education and learners do not share the same physical setting. Online learning should provide opportunities for learners to master necessary tasks. Online learners may use SRL strategies. In this research, we have collected, synthesized, and analyzed 130 articles on various topics related to SRL that published from 1986 to 2017, focusing on online learning and mathematics. We noted several models, phases, and few other topics discussed under SRL.

Adaptive mechanism for GA-NN to enhance prediction model

This research presents a hybrid Genetic Algorithm Neural Network (GA-NN) model to replace the physical tests procedures of Medium Density Fiberboard (MDF). Emphasis is on applying an adaptive mechanism on GA to enhance model performance. Data included in the model is MDF properties and its fiber characteristics. The focus of this study is the Multilayer Perceptron NN model, which is reliable to learn from seven inputs fed to the network to produce prediction of three targets. In order to avoid result from local optimum scenario, GA optimizes synaptic weights of the network towards reducing prediction error. The research used a fixed probability rates for crossover and mutation for hybrid GA-NN model. GA-NN model is further improved using adaptive mechanism to help identify the most suitable operator probability rates. The fitness value refers to Sum of Squared Error. Performance comparisons are between hybrid GA-NN and hybrid GA-NN with adaptive mechanism. Results show the hybrid GA-NN model with adaptive mechanism perform better than the ordinary hybrid model. The reliable model is able to simulate the testing procedure and therefore able to reduce the testing time required as well as to reduce the cost. Adaptive mechanism in GA helps increase capability to converge at zero sooner than the ordinary GA.

Embedding Dependability Attributes into Component-Based Software Development Using the Best Practice Method: A Guideline

Current organizational vulnerabilities mainly originate from Web applications. The security holes in Web applications have resulted in credit card theft, damaged financial resources and reputation of institutions, and compromised computers. Existing Web application systems encounter a high record of vulnerabilities that target dependability attributes. Mitigating software vulnerabilities and increasing software protection against bugs or vulnerabilities are critical to increase consumer confidence in software component products. Improved software engineering practices must also be adopted to mitigate the vulnerabilities in modern systems. Component-based software development (CBSD) is a software engineering approach. CBSD has been successfully applied in many domains. However, the CBSD capability to develop secure software applications is lacking to date. Therefore, this article proposes a guideline to overcome the lack of security trust in CBSD process. The proposed guideline embeds dependability attributes into CBSD by using the best practice method. The proposed guideline is significant for eliciting, analyzing, specifying, and composing the dependability attributes of CBSD.

Optimizing oil palm fiberboard properties using neural network

The shortage of rubber wood (RW) supply has increased the demand to reduce its composition in the Medium Density Fiberboard (MDF). Oil palm biomass such as empty fruit bunch (EFB) has been proven to be an excellent substitute to RW. An accurate percentage combination of RW and EFB will produce a high quality MDF. An MDF needs to be tested in terms of mechanical and physical properties so that it meets the required standard. These tests are costly for they involve high amount of resources. The aim of this research is to optimize the properties of MDF so that quality-testing procedures can be reduced. A prediction model will be used to make predictions on the MDF properties. A stepwise multiple linear regression selects the predictor variables to be used as inputs to the input nodes. With these variables, the multilayer perceptron neural network with various training criteria will train the data and finally produce the prediction. The results produced have shown that some of the property tests can be omitted.