Music Mood Detection Based On Audio And Lyrics With Deep Neural Net
This work addresses music mood detection for applications like music recommendation, though it appears incremental with comparisons to existing methods.
The researchers tackled multimodal music mood prediction using both audio signals and lyrics, showing their deep learning approach outperformed traditional feature engineering methods on arousal detection while performing equally on valence prediction, with their database containing 18,000 tracks.
We consider the task of multimodal music mood prediction based on the audio signal and the lyrics of a track. We reproduce the implementation of traditional feature engineering based approaches and propose a new model based on deep learning. We compare the performance of both approaches on a database containing 18,000 tracks with associated valence and arousal values and show that our approach outperforms classical models on the arousal detection task, and that both approaches perform equally on the valence prediction task. We also compare the a posteriori fusion with fusion of modalities optimized simultaneously with each unimodal model, and observe a significant improvement of valence prediction. We release part of our database for comparison purposes.