This repo contains an evaluated project carried out during the Data Science Lab course at Polito.
The assignment is to build a classification pipeline to predict the intent expressed in each audio recording in the dataset. More specifically, given an input audio sample, the goal is to predict both the action requested and the object affected by the action.