The Vocal Patterns Recognition In Artificial Neural Network By Using The Hebb Rule Algorithm

Abstract

ArtIficial network recognition Pattern Recognition is a branch of Artificial Intelligence that is better known as artificial neural networks. This method is growing rapidly at this time because it has a more accurate security system and has information that has a devised system that can be blocked with an attached system and embedded in a pattern recognition feature, such as pattern recognition that can be detected including signature patterns, facial patterns, and fingerprints. The problem that often occurs is the lack of information and securing data and confidentiality to produce information so that it can be used by authorized people only. The algorithm used in this study is the Hebb Rule algorithm. The Hebb Rule algorithm is an algorithm that has similarities to the McCulloch Pitts network architecture by using the concept of input units combined directly with output units, plus a bias value. The results of Pattern A = after being tested, the value is 4, Pattern I is recognized as -12, Pattern U is recognized as 13, Pattern E is recognized as much as 16, and Pattern O is recognized as much as 18. The final result of the research on recognizing vowels A, I, U, E, and O is unable to recognize patterns because they have different values in each vowel.