Klasifikasi Gender dan Usia Berdasarkan Suara dengan Menggunakan Metode KNN dan SVM

Speech recognition is one of the most important areas of research today, as it is used for a variety of applications such as security systems, authentication, human computer interaction, and psychological analysis. In this era of globalization, many technologies use human voices for their applications. As with communication over the phone, people are sometimes still confused to distinguish the voice of the person communicating over the phone. This research focuses on comparing the accuracy results of gender and age recognition classification based on voice using the K-Nearest Neighbor (KNN) method and the Support Vector Machine (SVM) method and using Mel Frequency Cepstral Coefficient (MFCC) feature extraction. Process analysis using python language. The data used is Biometrci Visions and Computing (BVC). The first test results from the BVC dataset with KNN classification resulted in the highest accuracy in gender 97.35% and age 88.73% at K = 3. For SVM classification, the highest accuracy for gender is 98.69% and age is 99.34% in the RBF kernel.