Music is one of the most popular forms of entertainment. Along with the development of information technology, music streaming platforms such as Spotify, Apple Music, and Deezer are increasingly popular among users. However, with thousands of songs available on these music streaming platforms, users often have difficulty finding songs that suit their tastes. Therefore, we design a music recommender system that can assist users in finding songs that are more in line with user preferences. In this research, we propose the development of a content-based music recommender system using a combination of Content-Based Filtering and Deep Neural Network (DNN) methods. The DNN used is Convolutional Neural Network (CNN) which serves to analyze the audio features of songs and learn user preferences to increase the percentage of accuracy in providing recommendations that match user needs. This recommender system works by extracting features from songs listened to by the user and then recommending other songs with similar features. We trained and evaluated our model on a dataset of 250 songs. This research aims to develop a music recommender system that can provide personalized recommendations to users according to the preferences of users. This research provides an accuracy result of 73.5%. From these results, it has been proven that the resulting music recommendations can be an alternative to the existing Collaborative Filtering-based recommender system.