 Hi everyone, thank you very much for coming for this tutorial on the drawing expression for this new project in Cuba. So, is it aware of the tutorial? The slides are available here, so you can read them out later on. So, this tutorial, we are really starting to mean that we will do the tutorial. We are KSC students here at the Ponte de Favre, and in my KSC Music Information Retrieval, and this is mainly KSC in natural language processing. So, this is mostly the joint work that we have been doing, we are all doing our KSC. It's not a comprehensive analysis of the state of the art in natural language processing in Miami. It's focused on the aspects we have worked on. We think that the most interesting areas know that there can be a intersection between these areas. We have the advice of all our pieces advisors, and we have a joint project here in the Ponte de Favre, that is called Music Meets LLB, where we have this tutorial that is supported by this project, and this is supported by Maria LLB. The objectives of this tutorial are, we'll provide a general introduction to LLB first, then try to address which areas of LLB might be interesting to use in Miami Art, and then address the extraction of semantic information from texts, and how to exploit this semantic information in Music Information Retrieval, and finally, we'll take some latest tendencies in natural language processing that can be also applied in Miami Art. So, why semantic information? So, in a typical Miami Art program, we can use the audio in Italy, the audio of the audio form, and use the audio for the task, the raw audio. Or we can get some low-level features from the audio, described in the audio, and use these features in our task, that can be machine learning, classification, whatever we want to do. So, we take these low-level features, or we can even extract high-level features from the audio. We can extract semantics like the mood, the genre, instruments, or high-level features, and use it in Music Information Retrieval task. So, this is a typical pipeline in Miami Art at the end. So, what happens with text? We can also use text in Miami Art. We can take the raw text and use it to count the words, and use these things in a Miami Art task. We can get surface features from the text, like part of a speech task, or other surface features, and use that for the task. So, there have been many papers in Miami Art using these approaches that use, like, vector speech model, part of speech task, concurrence, but we are not working on these tutorials, so this has to be done. So, what we propose is, okay, we can also get high-level features from text. In the same way that happens with audio, we can get semantics from the text, and then use these high-level semantic representations in a task. So, these semantic representations also come from the semantic web, and there have been some research in Miami Art also using semantic web as a representation in Miami Art. So, this tutorial is focused on these two aspects. One is how to get these high-level features from text, and how to exploit them in Miami Art. So, most of the research in Miami Art also dealing with text are based on lyrics corporate, and a few of them using other kinds of text, like biographies, blog formats, the videos, the libraries, and this tutorial is also focused on this task. We are not working with lyrics, we are working with biographies, we are writing the libraries, information that talks our music. And that's it. The outline of the talk is this. First, we are going to start with an introduction to NLP to define all the concepts in NLP in case you don't have experience with NLP before. Then we are going to focus more on information instruction that is a task in NLP. We have been working more on that, and then how to build music on the page, how to extract some of the information from text in another way, and then the application. We are going to apply that in Miami Art. We provide that in musicology. And then finally, we will talk about more latest tendencies, like lexical semantics and deep learning, and some conclassions at the end. So, let's start with this, that we will give this introduction to NLP.