Hussein Abdel Karim, the teaching instructor at the College of Education in Basra, participated in a scientific study within a research team from the universities of Saudi Arabia, Malaysia, and Nigeria. Shah researched a research paper in a solid scientific journal entitled “
Creating the Hu-Int Dataset: A Comprehensive Arabic Speech Dataset for Gender Detection and Age Estimation of Arabs”
"celebrities
The study, which was published in a prestigious scientific journal affiliated with the British Publishing House, was soon included in the repositories of Scopus Q1 and Clarvit. This scientific study will contribute to defining the requirements for creating a new data set for the most famous customers in the Arabic language.
The results of the study led to the design of the Hu-Int data set to identify the features of Arabs among themselves. It included 1,017 video clips after the filtering process and 93,814 video clips in different formats (such as mp4, wav and CSV). Seven algorithms were applied, namely Random Forest, Logistic, SVM (RBF), and SVM (Linear), then Tree and CNN-LSTM were used. Preliminary experimental results found that hybrid CNN-LSTM algorithms outperform many other algorithms in gender detection and age estimation.
Department of Media and Government Communication/College of Education.
