Automated Creation of Music using Deep Learning

Using the deep learning algorithms to make music without human input, automatic music generation has emerged as an exciting area of study This work offers a comprehensive overview of the advancements made in the area of deep learning algorithms-based automatic music production. The paper starts off by going over the fundamental ideas of deep learning and how they apply to the creation of music. Recurrent neural networks (RNN s), long short-term memory networks (LSTMs), and transformer-oriented models are a few examples of deep learning architectures frequently employed in music composition. The focus of the discussion shifts to the fundamental problems with automatic music production, including preserving musical coherence, preserving genre-specific characteristics, and resolving the problem of musical creativity and originality.