Gender Identification based on Speech Recognition using Backpropagation Neural Network
Laksita Maulisa Liztio, Christy Atika Sari, De Rosal Ignatius Moses Setiadi, Eko Hari Rachmawanto · 2020 International Seminar on Application for Technology of Information and Communication (iSemantic) · 2020
Research in the field of speech recognition is very interesting and challenging. Speech recognition can also be used to identify the gender of the owner. This research proposes the Backpropagation Neural Network (BPNN) method to recognize and classify gender based on sound as input. BPNN was chosen because it has the main characteristics that can learn nonlinear, input, and output complex and can adjust the data used. The experiments were carried out on three types of datasets namely private dataset, Kaggle dataset, and Javanese gender dataset. Based on the test results obtained the highest accuracy, which is 95%. Where out of the 100 voice datasets used, there was one data that failed to be identified and four others were incorrectly recognized.