Regresar

Classical music pbkp_rediction and composition by means of variational autoencoders

Abstract:

This paper proposes a new model for music pbkp_rediction based on Variational Autoencoders (VAEs). In thiswork, VAEs are used in a novelway to address two different issues: music representation into the latent space, and using this representation to make pbkp_redictions of the future note events of the musical piece. This approach was trained with different songs of Handel. As a result, the system can represent the music in the latent space, and make accurate pbkp_redictions. Therefore, the system can be used to compose new music either from an existing piece or from a random starting point. An additional feature of this system is that a small dataset was used for training. However, results show that the system is able to return accurate representations and pbkp_redictions on unseen data.