Game Belajar Membaca Kata Bahasa Inggris Dengan Pendekatan Metode MFCC dan KNN

Abstract

Learning English, especially in terms of speaking, at the elementary school level in rural areas with minimal technology is a complex challenge. In an era of globalization filled with technological advances, it is important for students to learn English from an early age. Internet and smartphone technologies, which are becoming increasingly prevalent in children's lives, can be used as innovative tools for learning English. A multimedia-based active and interactive learning approach can enhance the effectiveness of learning in this context. This research aims to develop innovative learning methods by focusing on creating internet and smartphone-based games to facilitate English speaking skills. These games use voice features extracted with the Mel Frequency Cepstral Coefficients (MFCC) method, which effectively recognizes differences in language and accent in speech recognition. The classification method employed is K-Nearest Neighbors (K-NN), chosen for its simplicity in implementation and suitability for datasets with many features. By integrating these technologies, the proposed approach can provide immediate feedback and engage students more effectively, making the learning process enjoyable and accessible, especially for those in rural areas. This innovation represents a significant step towards improving English education in resource-limited settings.