A Comprehensive Machine Learning Approach for Predicting Beats Per Minute (BPM) in Music Using Audio Features

Darsiti Darsiti

doi:10.65780/bima.v1i1.2

Authors

Darsiti Darsiti Universitas Teknologi Digital Author

DOI:

https://doi.org/10.65780/bima.v1i1.2

Keywords:

Beats Per Minute, Machine Learning, Audio Features, BPM Prediction, Ensemble Learning

Abstract

Predicting Beats Per Minute (BPM) in music is a significant challenge due to the complexity of the relationship between various audio features, such as rhythm, energy, and mood. Traditional methods are often unable to handle the complexity of feature variations and interactions. This study aims to develop a more accurate and reliable machine learning model to predict song BPM based on extracted audio features. We use advanced machine learning algorithms, including LightGBM, XGBoost, and Random Forest, to train models with a dataset covering ten audio features. Evaluation is performed using a k-fold cross-validation scheme with RMSE, MAE, and R² Score metrics. The experimental results show that boosting-based models such as LightGBM produce the best performance, with the lowest RMSE of 10.48, the lowest MAE of 7.62, and the highest R² Score of 0.83. However, these models still show a tendency to regress to the mean, indicating that some more extreme BPM variations are not fully captured. These findings emphasize the importance of improvements in feature engineering techniques and data rebalancing to improve BPM prediction accuracy in practical applications, such as music recommendation systems and tempo analysis.

Downloads

Download data is not yet available.

References

[1] M. Darvish and M. Bick, ‘The Role of Digital Technologies in the Music Industry—A Qualitative Trend Analysis’, Information Systems Management, vol. 41, no. 2, pp. 181–200, Apr. 2024, doi: 10.1080/10580530.2023.2225129.

[2] D. Newman, ‘Can the Key Signature, Mode, and the Beats per Minute of a Song Predict the Emotional Tone of Popular Music Lyrics?’, International Journal on Studies in English Language and Literature, vol. 11, p. 7, Dec. 2023, doi: 10.20431/2347-3134.1112002.

[3] S. Cole, ‘A song for each moment: Identifying listening modes as reflexive practices in music streaming - Sebastian Cole, 2025’, Accessed: Sep. 03, 2025. [Online]. Available: https://journals.sagepub.com/doi/full/10.1177/20539517251338742

[4] B. Amiri, N. Shahverdi, A. Haddadi, and Y. Ghahremani, ‘Beyond the Trends: Evolution and Future Directions in Music Recommender Systems Research’, IEEE Access, vol. 12, pp. 51500–51522, 2024, doi: 10.1109/ACCESS.2024.3386684.

[5] M. Mahat and D. Gracanin, ‘Deciphering Emotional Responses to Music: A Fusion of Psychophysiological Data Analysis and LSTM Predictive Modeling’, in Augmented Cognition, D. D. Schmorrow and C. M. Fidopiastis, Eds., Cham: Springer Nature Switzerland, 2024, pp. 52–67. doi: 10.1007/978-3-031-61569-6_4.

[6] D. B. Ramos, ‘Capturing Musical Prosody Through Interactive Audio/Visual Annotations’, phdthesis, Sorbonne Université, 2023. Accessed: Sep. 03, 2025. [Online]. Available: https://theses.hal.science/tel-04555575

[7] J. Zhang, S. Yu, R. Liu, G.-X. Xie, and L. Zurawicki, ‘Unveiling the melodic matrix: exploring genre-and-audio dynamics in the digital music popularity using machine learning techniques’, Marketing Intelligence & Planning, vol. 42, no. 8, pp. 1333–1352, Aug. 2024, doi: 10.1108/MIP-04-2024-0209.

[8] J. Mycka and J. Mańdziuk, ‘Artificial intelligence in music: recent trends and challenges’, Neural Comput & Applic, vol. 37, no. 2, pp. 801–839, Jan. 2025, doi: 10.1007/s00521-024-10555-x.

[9] T.-C. Pricop and A. Iftene, ‘Music Generation with Machine Learning and Deep Neural Networks’, Procedia Computer Science, vol. 246, pp. 1855–1864, Jan. 2024, doi: 10.1016/j.procs.2024.09.692.

[10] N. Sebastian, Jung, and F. Mayer, ‘Beyond Beats: A Recipe to Song Popularity? A machine learning approach’, 2024, arXiv. doi: 10.48550/ARXIV.2403.12079.

[11] M. Shu, ‘Exploring Spotify’s Music Popularity Dynamics and Forecasting with Machine Learning’, TNS, vol. 53, no. 1, pp. 83–89, Nov. 2024, doi: 10.54254/2753-8818/53/20240147.

[12] M. Walczyński and M. Kisz, ‘Using Machine Learning Algorithms to Explore Listeners Musical Tastes’, in Recent Challenges in Intelligent Information and Database Systems, vol. 1863, N. T. Nguyen, S. Boonsang, H. Fujita, B. Hnatkowska, T.-P. Hong, K. Pasupa, and A. Selamat, Eds., in Communications in Computer and Information Science, vol. 1863. , Cham: Springer Nature Switzerland, 2023, pp. 399–409. doi: 10.1007/978-3-031-42430-4_33.

[13] Z. Xing, ‘Popularity Prediction of Music by Machine Learning Models’, HSET, vol. 47, pp. 37–45, May 2023, doi: 10.54097/hset.v47i.8162.

[14] T. Jain, N. S. Kumar, V. Tiwari, and P. K. Tiwari, ‘Music Recommendation System Using Machine Learning’, in 2023 2nd International Conference on Futuristic Technologies (INCOFT), Belagavi, Karnataka, India: IEEE, Nov. 2023, pp. 1–6. doi: 10.1109/INCOFT60753.2023.10425522.

[15] X. Liu, ‘Music Trend Prediction Based on Improved LSTM and Random Forest Algorithm’, Journal of Sensors, vol. 2022, pp. 1–10, Mar. 2022, doi: 10.1155/2022/6450469.

[16] Konstantinos Gkountakos, Konstantinos Ioannidis, Konstantinos Demestichas, Stefanos Vrochidis, and Ioannis Kompatsiaris, ‘A Comprehensive Review of Deep Learning-based Anomaly Detection Methods for Precision Agriculture’, IEEE Access, pp. 1–1, 2024, doi: 10.1109/access.2024.3522248.

[17] S. Park, H. Sim, M. Kwon, and W. Lee, ‘Automatic Music Emotion Classification Using Chords and BPM’, in Future Information Technology, vol. 185, J. J. Park, L. T. Yang, and C. Lee, Eds., in Communications in Computer and Information Science, vol. 185. , Berlin, Heidelberg: Springer Berlin Heidelberg, 2011, pp. 329–336. doi: 10.1007/978-3-642-22309-9_42.

[18] A. C. North, A. E. Krause, L. P. Sheridan, and D. Ritchie, ‘Popularity, mood, energy, and typicality in music: A computerized analysis of 204,506 pieces.’, Psychology of Aesthetics, Creativity, and the Arts, vol. 13, no. 1, pp. 89–109, Feb. 2019, doi: 10.1037/aca0000165.

[19] J. Li and F. Yan, ‘Design and Application of Music Genre Classification Algorithm Based on Machine Learning’, in 2024 3rd International Conference on Data Analytics, Computing and Artificial Intelligence (ICDACAI), Zakopane, Poland: IEEE, Oct. 2024, pp. 1064–1068. doi: 10.1109/ICDACAI65086.2024.00198.

[20] R. Cavicchioli, J. C. Hu, and M. Furini, ‘On Using Artificial Intelligence to Predict Music Playlist Success’, in 2024 IEEE 21st Consumer Communications & Networking Conference (CCNC), Las Vegas, NV, USA: IEEE, Jan. 2024, pp. 278–283. doi: 10.1109/CCNC51664.2024.10454829.

[21] A. Natev, ‘Hit Song Prediction Through Machine Learning and Spotify Data’, in Proceedings of the 10th Student Computing Research Symposium (SCORES’24), University of Maribor Press, Oct. 2024, pp. 57–60. doi: 10.18690/um.feri.6.2024.13.

[22] I. Olvera, K. Sood, C. Reyes, and W. Tu, ‘Content-Based Music Recommendation System using Supervised Learning’, in 2023 IEEE International Conference on Communication, Networks and Satellite (COMNETSAT), Malang, Indonesia: IEEE, Nov. 2023, pp. 120–127. doi: 10.1109/COMNETSAT59769.2023.10420756.

[23] S. Arora, R. Rani, and N. Saxena, ‘Music Stream Analysis for the Prediction of Song Popularity using Machine Learning and Deep Learning Approach’, in 2022 3rd International Conference on Computing, Analytics and Networks (ICAN), Rajpura, Punjab, India: IEEE, Nov. 2022, pp. 1–5. doi: 10.1109/ICAN56228.2022.10006843.