This project has started in 2019 with the intention to understand what are the approaches for:

• Music Analytics - What are the factors that make us define a music in a certain way? Genre, likebility, etc. What are those features and how can use computational methods in order to extract them?

• Generative Learning - WIP with some drafts at this Github repo. How can we use machine learning merge well with arts and vice versa? How technology can help us to become more creative and how artists can use technology as and id for their work? These are some questions that this project will try to answer. Technology has been know to be straightforward, understand and replicate rules. However, it's also know to be able to process and analyse waves of data and provide a smart way for us humans to access it. Considering these both aspects, why can't machines search in the vast amount of data and help us to understand how certain musicians played? What were their preferred music motifs? How did he like to play? Can we even quantify this? In the past recent years, the field of Generative Learning has researched how the most advanced machine learning models can understand, replicate and modify abstract art models, from painting, to music composition and game playing. Here I want to investigate the major advances in Deep Learning and how they manage to achieve that. From CNNs, LSTMs to Autoenconders and LOTS of GANS, I'll try to replicate and extend some examples given in the Generative Deep Learning book