Using High Dimensional Computing on Arabic Language Speech to Text Classification

Authors

George S. Rady, Department of Computers and Information Technology, The Egyptian E-Learning University, EgyptFollow
Mamdouh F. Mohamed, Department of Computer Science, Faculty of Computers and Information, Assiut University, EgyptFollow
Khaled F. Hussain, Department of Computer Science, Faculty of Computers and Information, Assiut University, EgyptFollow

Abstract

High-Dimensional Processing is the idea that mind register illustrations of neural activities which are not immediately related with numbers. The objective of the article is hyper- dimensional computation of data for categorization of text from two distinct speech datasets, namely the Arabic Corpus dataset and the MediaSpeech dataset with four languages (Arabic, Spanish, French, and Turkish). Through the use of an n-gram encoding scheme, hyper dimensional computing is used to conduct the analysis from the prior set of data. Using hyper dimensional computing, the MediaSpeech dataset accomplishes 100% accuracy for all 4-gram to 14-gram encoding schemes, while the Arabic Corpus dataset accomplishes 100% accuracy for 4-gram to 7-gram encoding schemes.

Recommended Citation

S. Rady, George; F. Mohamed, Mamdouh; and F. Hussain, Khaled (2023) "Using High Dimensional Computing on Arabic Language Speech to Text Classification," Information Sciences Letters: Vol. 12 : Iss. 3 , PP -.
Available at: https://digitalcommons.aaru.edu.jo/isl/vol12/iss3/6