AttAE-RL²: Attention based Autoencoder for Rap Lyrics Representation Learning

Hongru Liang, Qian Li, Haozheng Wang, Hang Li, Jin-Mao Wei, Zhenglu Yang

Abstract: Learning rap lyrics is an important area of music information retrieval because it is the basis of many applications, such as recommendation systems, automatic classification. In this paper, we tackle the issue pertaining to the lack of an effective approach to aggregate various features of lyrics by proposing an attention-based autoencoder for rap lyrics representation learning (AttAE-RL²). The proposed method appropriately integrates the semantic and prosodic features of rap lyrics. The preliminary experimental results demonstrate that our approach outperforms the state-of-the-art ones.