In this study, the authors propose a speech recognition system using harmonic structure related information to detect harmonic features in noisy environment. The proposed algorithm first extracts the harmonic components contained inside the speech signals using sine function convolution. By setting the frequency of the sine function as equal to the fundamental frequency of speech signals, harmonic components can be extracted out. The reconstructed signal obtained by summing up the extracted harmonic components is found to have a high degree of correlation with the original signal. The extracted frame energy measure of the harmonic components has been further processed to become dynamic harmonic features and then used together with the European Telecommunications Standards Institute (ETSI) front-end processed mel-frequency cepstral coefficients (MFCC) feature or the perceptual linear prediction (PLP) feature in the speech recognition system. The proposed enhanced speech recognition system shows a better recognition rate over the ETSI front-end processed MFCC (or PLP)-based speech recognition system.

No abstract

No abstract

Kalman filter is normally used to enhance speech quality in a noisy environment, in which the speech signals are usually modelled as autoregressive (AR) process, and represented in the state-space domain. It is a known fact that to identify the changing AR coefficients in every time state requires extensive computation. In this paper, the authors develop a bidirectional Kalman filter and apply it in a speech processing system. The proposed filter uses a system dynamics model that utilises the past and the future measurements to form an estimate of the system's current time state. It provides efficient recursive means to estimate the state of a process that minimises the mean of the squared error. Compared to the conventional Kalman filter, the proposed filter reduces the computation time in two ways: (i) by avoiding the computation of AR parameters in each time state, and (ii) by reducing the dimension of the matrices involved in the difference equations and the measurement equations into constant (1 × 1) matrices. The speech recognition result shows that the developed speech recognition system becomes more robust after the proposed filtering process, and the proposed filter's low computational expense makes it applicable in the practical hidden Markov model-based speech recognition system.

Schooling system must provide high quality learning opportunities to meet the educational needs and ensuring achievement for every student. All teachers monitor their students’ progress throughout the year, includes formative assessment, questioning, providing feedback, etc. This practice helps teachers continually assess students’ academic performance and evaluate the effectiveness of their teaching. In this paper, k-means clustering method with deterministic model is used to analyze the student's overall performance. The results is important for educators to identify students who are at risk academically and areas where teaching strategies may need adjustment to better meet these students' needs.

Sign language is a form of communication language designed to link a deaf-mute person to the world. To express an idea it requires the use of hand gestures and body movement. However, the bulk of the general population remain uneducated to understand the sign language. Therefore, a translator is required to facilitate the communication. This paper wishes to extend the previously proposed Convolutional Neural Network (CNN) model for predicting American Sign Language with a MobileNetV2-based transfer learning model. The latter model effectively generalized on a dataset which is around 18 times larger with 5 additional groups of hand signs. Over 98% of the recognition accuracy had been reported. Because of its relatively fewer parameters and less intensive computational operations compared to other deep learning architectures, the model was also ideal to be implemented on mobile devices. The model will serve as the key to deploying a sign language translator software on smartphone to enhance communication efficiency between the deaf-mute person and the general public.

No abstract

No abstract