A Neural Network for Composer Classification
Gianluca Micchi · HAL (Le Centre pour la Communication Scientifique Directe) · 2018
I present a neural network approach to automatically extract musical features from 20-second audio clips in order to predict their composer. The network is composed of three convolutional layers followed by a long short-term memory recurrent layer. The model reaches an accuracy of 70% on the validation set when classifying amongst 6 composers. The work represents the early stage of a project devoted to automatic feature detection and visualization.