Browsing by Author "Reyal, S"

Now showing 1 - 12 of 12

Embargo
Adapting MaryTTS for Synthesizing Sinhalese Speech to Communicate with Children
(IEEE, 2021-12-01) Lakmal, M. A. J. A; Methmini, K. A. D. G; Rupasinghe, D. M. H. M; Hettiarachchi, D. I; Piyawardana, V; Senarathna, M; Reyal, S; Pulasinghe, K
The majority of the Sri Lankan population speak Sinhala, which is also the country's mother tongue. Sinhala is a difficult language to learn by children aged between 1–6 years when compared to other languages. Text to speech system is popular among children who have difficulties with reading, especially those who struggle with decoding. By presenting the words auditorily, the child can focus on the meaning of words instead of spending all their brainpower trying to sound out the words. In Sri Lanka, however, computer systems based on the Sinhala language especially for children are extremely rare. In this situation having a Sinhala text-to-speech technology for communicating with children is a helpful option. Intelligibility should be considered deeply in this system because this is specific for children. Recordings of a native Sinhalese speaker were used to synthesize a natural-sounding voice, rather than a robotic voice. This paper proposes an approach of implementing a Sinhalese text-to-speech system for communicating with children using unit selection and HMM -based mechanisms in the MaryTTS framework. Although a work in progress, the intermediate findings have been presented.
Embargo
Adaptivo: A Personalized Adaptive E-Learning System based on Learning Styles and Prior Knowledge
(IEEE, 2022-12-09) Rishard, M.A.M; Jayasekara, S.L; Ekanayake, E.M.P.U; Wickramathilake, K.M.J.S; Reyal, S; Manathunga, K; Wickramarathne, J
The rapid advancement of technology and the internet has resulted in an increase in the number of learners seeking e-learning. Though E-Learning is widely used most e-learning systems provide the same set of learning resources and learning paths to each student, regardless of their personal preferences. In recent years there has been increasing attention towards the characteristics of learners such as the learning styles and the knowledge level of the learner. This research paper proposes a personalized adaptive E-learning system called “Adaptivo” that provides a personalized learning experience to the learners based on their learning style and knowledge level. To make the learning process more efficient and engaging, Adaptivo takes into account the specific differences between learners in terms of time, online interactions and learning duration. It then builds a personalized learning path depending on each learner's learning style and knowledge level. The main aim of this study is to investigate the impact of the proposed adaptive learning approach on learners. The results show that the students appreciate the approach, are highly satisfied, and performed better when content is personalized according to their learning style and prior knowledge.
Embargo
Artificial Intelligence Based Smart Library Management System
(IEEE, 2021-12-01) Jayawardena, C; Reyal, S; Kekirideniya, K. R; Wijayawardhana, G. H. T; Rupasinghe, D. G. I. U; Lakranda, S. Y. R. M
The concept of a smart library system is to operate in a library with minimal human intervention. The proposed system handles basic library functions such as issuing books, returning books, reserving books, collecting late return fines, antitheft detection, and managing booking inventory using IoT technologies such as RFID and Raspberry Pi. The first key criterion is shelves management, which provides a unique key and uses RFID technology to identify and arrange the books. The succeeding category is providing users with book recommendations by penalizing hidden layer activations to encourage only a few nodes to activate when a single sample is an input. The dimensionality reduction neural network method was used to select the optimal seating arrangement. The LSTM algorithm will be used to make predictions to provide an efficient service to library users.
Embargo
Automated Programming Assignment Marking Tool
(IEEE, 2022-07-18) Thenuwara, T. B. K. P; Vimalaraj, H; Wijekoon, V. U; Sathurjan, T; Reyal, S; Kuruppu, T. A; Tharmaseelan, J
Due to the enrolment of a very high number of students to programming modules, marking of programming modules is becoming a very tedious and time-consuming process. Programming assignments mainly test for the student’s ability to think logically and approach a solution to the problem. In that case, just running the script and checking the output will not be sufficient enough to award a grade to the student. Marking criteria of programming modules provide certain marks for programs which are not syntactically correct but still have a good approach. Therefore, the code has to be read line by line and the implementation should be checked carefully to provide marks. Source code analysis has become mandatory in the current scenario. This leads to immense pressure and heavy workload on the staff who mark these programs. Considering all these aspects manual marking can lead to inconsistency, biasness, waste of time and less accuracy. Therefore, the main objective of this research is to minimize these problems by implementing an automated programming module marking tool by converting source codes to parse trees, extracting features, generating feature vectors, comparing them and generating a mark along with a feedback and plagiarism report. The solution focuses on automation marking by source code analysis and plagiarism checking.
Embargo
Automated Programming Assignment Marking Tool
(IEEE, 2022-07-18) Vimalaraj, H; Thenuwara, T. B. K. P.; Wijekoon, V. U; Sathurjan, T; Reyal, S; Kuruppu, T. A; Tharmaseelan, J
Due to the enrolment of a very high number of students to programming modules, marking of programming modules is becoming a very tedious and time-consuming process. Programming assignments mainly test for the student’s ability to think logically and approach a solution to the problem. In that case, just running the script and checking the output will not be sufficient enough to award a grade to the student. Marking criteria of programming modules provide certain marks for programs which are not syntactically correct but still have a good approach. Therefore, the code has to be read line by line and the implementation should be checked carefully to provide marks. Source code analysis has become mandatory in the current scenario. This leads to immense pressure and heavy workload on the staff who mark these programs. Considering all these aspects manual marking can lead to inconsistency, biasness, waste of time and less accuracy. Therefore, the main objective of this research is to minimize these problems by implementing an automated programming module marking tool by converting source codes to parse trees, extracting features, generating feature vectors, comparing them and generating a mark along with a feedback and plagiarism report. The solution focuses on automation marking by source code analysis and plagiarism checking.
Embargo
Comparative Study of Parameter Selection for Enhanced Edge Inference for a Multi-Output Regression model for Head Pose Estimation
(Institute of Electrical and Electronics Engineers Inc., 2022-11-04) Lindamulage, A; Kodagoda, N; Reyal, S; Samarasinghe, P; Yogarajah, P
Magnitude-based pruning is a technique used to optimise deep learning models for edge inference. We have achieved over 75% model size reduction with a higher accuracy than the original multi-output regression model for head-pose estimation
Embargo
COVID-Tracker: Surveillance of Potential Clusters Using a Wristband and Location-based Data
(IEEE, 2022-07-18) Mandara, A.P. M; Randula, H.K. K; Priyadarshana, H. L.Y; Uyanahewa, J. J.; Manathunga, K; Reyal, S
COVID-19 is a global pandemic that has threatened the survival of humans and other living beings. COVID-19 causes illnesses varying from the very mild cold to serious health complications resulting in death. Most Information Technology based solutions have been implemented to prevent the COVID-19 pandemic while raising awareness in the public. However, there is a limited number of reliable and real-time applications of self-awareness on COVID-19. Currently, the globe is dealing with the COVID-19 epidemic, particularly in pursuit of economic growth in each country. Therefore, an accurate, efficient automatic method to raise self-awareness by avoiding risky contacts is useful for human survival. This paper describes the automatic detection of temperature using a wearable device and an automatic alerting mechanism to inform the users of potentially risky contacts with higher temperatures nearby within a considerable time frame. COVID-Tracker produces results with high accuracy and efficiency, this is beneficial to improve self-awareness among users, to visualize potential covid clusters, and also to improve the mental health of self-isolated people. The developed application consists of four main components namely: temperature measuring band, mobile application, prediction model-based visualization dashboard and an AI bot. Based on the results reported here, developed methods can help people to achieve self-awareness of COVID-19 by avoiding risk factors early and accurately.
Embargo
Elegant Fit-On – Virtual Fitting Room on Handheld Devices
(IEEE, 2022-12-09) Galagoda, R.R.N.P.A.B.W.M.S.R; Gunarathne, E.H.N.L.; Maheshi Purnima, K.A.D.; Wickramarathna, H.P.C.S.; Reyal, S; Siriwardana, S
Clothing has been one of the basic human needs since ancient times. It is a common thing to try on clothes and consider certain features when buying clothes. With the current pandemic situation, it is risky to wear and buy clothes by physical shopping. Consequently, people do online shopping. Those existing shopping websites are not user-friendly and less reliable as the customers will not have the privilege to purchase the exactly fitting outfit. Therefore, the customer satisfaction level is low with the clothes they have bought through online platforms. Therefore, the aim is to utilize technology to provide a virtual fitting room experience on handheld devices. The objective is to create a customized 3D avatar that represents the customer’s unique body shapes and features, which allows to try on clothes. This avatar is 360 degrees rotatable with pre-defined poses to check what the fit-on looks like. This solution shows whether the clothes are too fit or loose for the customer by showing live wrinkles. The text and voice feedback are generated at the end, which would be helpful for differently-abled people, especially those with vision issues.
Embargo
Revisit of Automated Marking Techniques for Programming Assignments
(IEEE, 2021-04-21) Tharmaseelan, J; Manathunga, K; Reyal, S; Kasthurirathna, D; Thurairasa, T
Due to the popularity of the Computer science field many students study programming. With large numbers of student enrollments in undergraduate courses, assessing programming submissions is becoming an increasingly tedious task that requires high cognitive load, and considerable amount of time and effort. Programming assignments usually contain algorithmic implementations written in specific programming languages to assess students' logical thinking and problem-solving skills. Evaluators use either a test case-driven or source code analysis approach when evaluating programming assignments. Given that many marking rubrics and evaluation criteria provide partial marks for programs that are not syntactically correct, evaluators are required to analyze the source code during evaluations. This extra step adds additional burden on evaluators that consumes more time and effort. Hence, this research work attempts to study existing automatic source code analysis mechanisms, specifically, use of deep learning approaches in the domain of automatic assessments. Such knowledge may lead to creating novel automated marking models using past student data and apply deep learning techniques to implement automatic assessments of programming assignments irrespective of the computer language or the algorithm implemented.
Open Access
Source Code based Approaches to Automate Marking in Programming Assignments
(Science and Technology Publications, 2021) Kuruppu, T; Tharmaseelan, J; Silva, C; Samaratunge Arachchillage, U. S. S; Manathunga, K; Reyal, S; Kodagoda, N; Jayalath, T
With the embarkment of this technological era, a significant demand over programming modules can be observed among university students in larger volume. When figures grow exponentially, manual assessments and evaluations would be a tedious and error-prone activity, thus marking automation has become fast growing necessity. To fulfil this objective, in this review paper, authors present literature on automated assessment of coding exercises, analyse the literature from four dimensions as Machine Learning approaches, Source Graph Generation, Domain Specific Languages, and Static Code Analysis. These approaches are reviewed on three main aspects: accuracy, efficiency, and user-experience. The paper finally describes a series of recommendations for standardizing the evaluation and benchmarking of marking automation tools for future researchers to obtain a strong empirical footing on the domain, thereby leading to further advancements in the field.
Open Access
Source Code based Approaches to Automate Marking in Programming Assignments.
(Science and Technology Publications, 2021) Kuruppu, T; Tharmaseelan, J; Silva, C; Samaratunge Arachchillage, U. S. S; Manathunga, K; Reyal, S; Kodagoda, N
With the embarkment of this technological era, a significant demand over programming modules can be observed among university students in larger volume. When figures grow exponentially, manual assessments and evaluations would be a tedious and error-prone activity, thus marking automation has become fast growing necessity. To fulfil this objective, in this review paper, authors present literature on automated assessment of coding exercises, analyse the literature from four dimensions as Machine Learning approaches, Source Graph Generation, Domain Specific Languages, and Static Code Analysis. These approaches are reviewed on three main aspects: accuracy, efficiency, and user-experience. The paper finally describes a series of recommendations for standardizing the evaluation and benchmarking of marking automation tools for future researchers to obtain a strong empirical footing on the domain, thereby leading to further advancements in the field.
Embargo
Step-by-Step Process of Building Voices for Under Resourced Languages using MARY TTS Platform
(IEEE, 2022-12-09) Senarathna, M; Pulasinghe, K; Reyal, S
This paper presents a comprehensive guide for creating synthetic voices to support under resourced languages for the MaryTTS platform. Although researchers have extensively contributed in the domain of speech synthesis, the lack of a thorough documentation hinders the voice building process for languages not yet supported by MaryTTS, complicating the implementation process for users with inadequate knowledge in the field of Text-to-Speech (TTS). The step-by-step process discussed in this study is further demonstrated with the creation of a synthetic voice for the Sinhala language, with unit selection as the voice building approach. A Sinhalese voice was generated with an intelligibility score of 91.7% upon evaluation with Diagnostic Rhyme Test (DRT). Comparison with ground truth data proved a close approximation to human speech where the intelligibility score was identified as 97.9%, when tested with the same participants. The Mean Opinion Score (MOS) revealed a naturalness level of 2.993, indicating a moderately high speech quality for the proposed system in comparison with the ideal score of 4.972.