# Music Mood Detection Based On Audio And Lyrics With Deep Neural Net

arXiv: https://arxiv.org/abs/1809.07276 · HTML: https://mlsanity.com/paper/1809.07276/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Sep 2018.

## Built on by (most cited first, 14 of 14)

- [Contrastive Audio-Language Learning for Music](https://mlsanity.com/paper/2208.12208/) (2022)
- [Cold Start Similar Artists Ranking with Gravity-Inspired Graph Autoencoders](https://mlsanity.com/paper/2108.01053/) (2021)
- [Flow Moods: Recommending Music by Moods on Deezer](https://mlsanity.com/paper/2207.11229/) (2022)
- [Are We There Yet? A Brief Survey of Music Emotion Prediction Datasets, Models and Outstanding Challenges](https://mlsanity.com/paper/2406.08809/) (2024)
- [Interpreting Song Lyrics with an Audio-Informed Pre-trained Language Model](https://mlsanity.com/paper/2208.11671/) (2022)
- [MERGE -- A Bimodal Audio-Lyrics Dataset for Static Music Emotion Recognition](https://mlsanity.com/paper/2407.06060/) (2024)
- [MusicLIME: Explainable Multimodal Music Understanding](https://mlsanity.com/paper/2409.10496/) (2024)
- [Lyric document embeddings for music tagging](https://mlsanity.com/paper/2112.11436/) (2021)
- [Multi-Modality in Music: Predicting Emotion in Music from High-Level Audio Features and Lyrics](https://mlsanity.com/paper/2302.13321/) (2023)
- [A Computational Analysis of Lyric Similarity Perception](https://mlsanity.com/paper/2404.02342/) (2024)
- [The Contribution of Lyrics and Acoustics to Collaborative Understanding of Mood](https://mlsanity.com/paper/2207.05680/) (2022)
- [LyCon: Lyrics Reconstruction from the Bag-of-Words Using Large Language Models](https://mlsanity.com/paper/2408.14750/) (2024)
- [Semi-Supervised Self-Learning Enhanced Music Emotion Recognition](https://mlsanity.com/paper/2410.21897/) (2024)
- [Enhancing Stroke Diagnosis in the Brain Using a Weighted Deep Learning Approach](https://mlsanity.com/paper/2504.13974/) (2025)

## Abstract (arXiv)

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.