ABSTRACTIn this research, we delve into the identification of human voices based kata.on gender by leveraging the differences in vocal characteristics between males and females. In addition to differences in vocal tract size, factors such as length, thickness, and vocal cord stiffness also play a role in producing differences in the fundamental frequency of voice between the two genders. The fundamental frequency of voice becomes one of the indicators used in acoustic analysis for gender classification based on voice. In the automatic classification of voices, sound processing techniques and machine learning are pivotal in system development. The method of gender recognition based on voice involves acoustic analysis using voice features such as fundamental frequency, formants, duration, intensity, and intonation patterns. The research yielded an accuracy of 92% through modeling using CNN on audio data, and the testing results were quite satisfactory in terms of classification. This model's results have been implemented into a Flask API, serving as a connection or backend for an application. The application takes the form of a movie recommendation system developed using the Flutter framework. Consequently, within the movie application, there is voice clustering or classification of user voices to provide film recommendations within the application


Keywords – Deep Learning, Voice Recognition, Audio Classification, CNN, Gender