Music Classification in MIDI Format based on LSTM Mdel
This work addresses music classification for distinguishing AI from human composers, but it is incremental as it applies existing deep learning methods to a new dataset.
The paper tackled the problem of classifying music as AI-generated or human-composed by transforming MIDI files into natural language sequences and using an mLSTM model with logistic regression, achieving 90% accuracy in 10-fold cross-validation.
Music classification between music made by AI or human composers can be done by deep learning networks. We first transformed music samples in midi format to natural language sequences, then classified these samples by mLSTM (multiplicative Long Short Term Memory) + logistic regression. The accuracy of the result evaluated by 10-fold cross validation can reach 90%. Our work indicates that music generated by AI and human composers do have different characteristics, which can be learned by deep learning networks.