Intelligent Signal Processing Group (ISP)
http://dspace.unimap.edu.my:80/xmlui/handle/123456789/38870
2024-03-28T16:18:04ZHeterogeneous speech prediction using LDA classifiers
http://dspace.unimap.edu.my:80/xmlui/handle/123456789/38995
Heterogeneous speech prediction using LDA classifiers
Mohd Ridzwan, Tamjis; Muhammad Naufal, Mansor; Ahmad Kadri, Junoh; Amran, Ahmed; Wan Suhana, Wan Daud; Azrini, Idris
Classroom speech intelligibility has become one of the main concerns in schools and other learning institutions development nowadays. This is because the qualities of student’s perceptions towards teacher are essentials in learning development. Measures have been introduced by the acoustical association to tackle the speech intelligibility problems in the classroom such as room renovations. Room’s acoustics standards have been introduced in several countries but still the questions on whether the standards fits on every classroom in different countries are still arise. Studies have also shown that most of the researches that have been conducted were only focusing on the conventional type classroom which depends only on the teacher’s vocal power. This paper will formulate the measurement protocol on measuring the speech intelligibility in the sound reinforced (multiple speaker) classroom. Finally it was found that the speech intelligibility in the sound reinforced classroom is better than the conventional classroom by using Linear Discriminant Analysis.
Link to publisher's homepage at http://www.ttp.net/
2014-08-01T00:00:00ZInfant pain detection with homomorphic filter and fuzzy k-NN classifier
http://dspace.unimap.edu.my:80/xmlui/handle/123456789/38989
Infant pain detection with homomorphic filter and fuzzy k-NN classifier
Muhammad Naufal, Mansor; Ahmad Kadri, Junoh; Amran, Ahmed; Kamarudin, Hussin, Brig. Jen. Dato' Prof. Dr.; Azrini, Idris
Newborn pain is a non-stationary made by babies in reaction to certain circumstances. This infant facial expression can be used to recognize physical or psychology condition of newborn. The goal of this study is to evaluate the performance of illumination levels for infant pain classification. Local Binary Pattern (LBP) features are computed at Fuzzy k-NN classifier. Eight different performance measurements such as Sensitivity, Specificity, Accuracy, Area under Curve (AUC), Cohen's kappa (k), Precession, F-Measure and Time Consumption are performed. Fuzzy k-NN classifier is employed to classify the newborn pain. The outcomes accentuated that the suggested features and classification algorithms can be employed to assist the medical professionals for diagnosing pathological condition of newborn pain.
Link to publisher's homepage at http://www.ttp.net/
2014-08-01T00:00:00Z