LLM’s for Spanish Song Text Analysis and Classification Using Language Variants

Feelings are the affective state of mind, which are produced in the brain and are caused by an emotion. This feelings have been transferred in multiple ways, such as texts, paintings music. Music transmits different emotions, which makes it even more important to know what kind of feelings are found within a song, which can be, among many others, positive, negative or neutral. Through this work, the texts of two different datasets of songs in the Spanish language were analyzed and classified, this through BERT, RoBERTa and DistilBERT. These experiments showed an improvement compared with previous works and improvement with the use of these methods.