Deep learning based music genre classification algorithm
Tang He-ming, Fucheng You, Yuqin Liu · 2023
The classification of music genres is of great significance in the research and applications related to the efficient organization, retrieval and recommendation of music resources. To address the current problem of low accuracy of digital music genre classification algorithms based on deep learning, this paper improves the ECAPA-TDNN model proposed based on the x-vector architecture of TDNN by implanting a bi-directional LSTM network to obtain more information about temporal context, based on which the ECAPA-TDNN-BLSTM model. Finally, a dataset containing 1000 music tracks from ten music genres was divided and tested, and the effectiveness of the model was proved by the good results in the test set. The results show that the accuracy of the proposed model for music genre classification and recognition has been improved compared with the existing models.