African Beats and Lyrics Composition Using Recurrent Neural Networks and Transformer-Based Models
Authors
Computer Science Department, Caleb University, Imota, Lagos. (Nigeria)
Computer Science Department, Babcock University, Ilisan-remo, Ogun. (Nigeria)
Computer Science Department, Anchor University, Ipaja, Lagos (Nigeria)
Article Information
DOI: 10.51244/IJRSI.2026.1309000100
Subject Category: Computer Science
Volume/Issue: 13/9 | Page No: 1391-1408
Publication Timeline
Submitted: 2026-09-21
Accepted: 2026-09-26
Published: 2026-10-09
Abstract
Human music composition frequently struggles with challenges such as ensuring originality, maintaining lyrical coherence, and managing elaborate polyrhythms. These difficulties are particularly evident in African music, which is characterized by complex rhythms, call and response structures, rhythmic improvisation, and tonal language. This study explores how artificial intelligence can support and enhance African music composition by integrating Recurrent Neural Networks (RNNs) and transformer based models. The hybrid system was designed to generate both beats and lyrics, drawing on pre trained models such as Mistral 7B and MusicGen. The methodology adopted includes model pre training, data preparation, adaptation, composition pipelining, inference generation, post processing, and evaluation. Results demonstrated that the hybrid model achieved improved rhythmic consistency, lyrical coherence, and Afrobeat authenticity. Quantitative evaluation indicated lower error rates and stronger stylistic regularity, while subjective listener assessments averaged 4.29/5, indicating positive reception. When compared with existing models, the hybrid approach balanced structural coherence with creative diversity, offering a culturally grounded framework for AI driven African music composition.
Keywords
African beats, Afro-beats, Beats and lyrics generator, Hybrid model, rnn+transformer model.
Downloads
References
1. Adebayo, T., Okonkwo, C., & Mensah, J. (2024). Hybrid neural architectures for African beat generation and lyric synchronization. Journal of Computational Musicology, 12(3), 45–59. [Google Scholar] [Crossref]
2. Adekunle, O., Chinwe, E., & Bello, R. (2025). Multilingual transformer models for African lyric generation. IEEE Transactions on Affective Computing, 16(2), 210–223. [Google Scholar] [Crossref]
3. Ajilore, O. O., Eludire, A., Olumoye, M. Y., Adegunwa, O., Omotayo, O., Adekunle, A., & Olumoye, Y. (2023). Image classification algorithms for early detection of learning disabilities using visual data. International Journal of Innovative Science and Research Technology, 8(11), 815-819. [Google Scholar] [Crossref]
4. Ajilore, O. O., Phillip, A., Olumoye, M.Y., Eludire, A., Akanni, A., Adegunwa, O., (2025). Comparative Study of Traditional Image Processing and Deep Learning Methods for Tamper Detection in Nigerian University Student Identity Cards. International Journal of Computer Science and Information Security 19 (4), 109-126 [Google Scholar] [Crossref]
5. Briot, J.-P., & Pachet, F. (2020). Deep learning techniques for music generation: A survey. ACM Computing Surveys, 52(3), 1–34. [Google Scholar] [Crossref]
6. Brown T, Mann B, Ryder N, Subbiah M, Kaplan JD, Dhariwal P, Neelakantan A, Shyam P, Sastry G, Askell A, Agarwal S. (2020). Language models are few-shot learners. Advances in Neural Information Processing Systems, 33, 1877–1901. [Google Scholar] [Crossref]
7. Chen, Y., Zhang, L., & Li, H. (2022). Deep learning for symbolic music generation: A comparative study of RNN and transformer models. Neural Computing and Applications, 34(15), 12689–12703. [Google Scholar] [Crossref]
8. Gao, S., & Bailey, J. (2025). Large-scale language models for creative music composition. Artificial Intelligence Review, 58(2), 1123–1145. [Google Scholar] [Crossref]
9. Gadotti, A., Silva, R., & Pereira, L. (2024). Secure hashing techniques for anonymized datasets. Journal of Information Security, 13(1), 77–89. [Google Scholar] [Crossref]
10. Ghoshal, S., Chowdhury, P., & Buckchash, H. (2026). Fusing memory and attention: A study on LSTM, transformer and hybrid architectures for symbolic music generation. arXiv preprint arXiv:2602.19816. [Google Scholar] [Crossref]
11. Hirschprung, R., & Leshman, D. (2021). Data anonymization in AI systems: Techniques and challenges. Journal of Privacy and Data Protection, 5(2), 101–118. [Google Scholar] [Crossref]
12. Huang, A., Vaswani, A., Uszkoreit, J., Shazeer, N., Simon, I., & Eck, D. (2019). Music transformer: Generating music with long-term structure. International Conference on Learning Representations (ICLR). [Google Scholar] [Crossref]
13. Laidlow, J. (2024). Ethical considerations in AI-driven music composition. AI & Society, 39(4), 765–778. [Google Scholar] [Crossref]
14. Lison, P., Tiedemann, J., & Kouylekov, M. (2021). Data masking for privacy-preserving NLP. Computational Linguistics, 47(2), 345–362. [Google Scholar] [Crossref]
15. Liu, Y., Wang, H., & Chen, Z. (2024). Copyright compliance in AI-generated music datasets. Journal of Intellectual Property Law & Practice, 19(3), 215–229. [Google Scholar] [Crossref]
16. Mensah, J., & Okonkwo, C. (2025). Attention-driven lyric modeling for African languages using Transformer-XL. Applied Artificial Intelligence Review, 9(1), 88–104. [Google Scholar] [Crossref]
17. Naseem, M., Hussain, S., & Khan, A. (2021). Computational creativity in music composition: Challenges and opportunities. Journal of Creative Technologies, 15(2), 33–47. [Google Scholar] [Crossref]
18. Osei, K., Adebayo, T., & Ndlovu, S. (2024). Ethnomusicological priors in neural music composition: African rhythmic structures and AI. Computational Creativity Journal, 7(4), 33–47. [Google Scholar] [Crossref]
19. Payne, C. (2020). MuseNet and the evolution of generative music systems. OpenAI Technical Report. [Google Scholar] [Crossref]
20. Schmidhuber, J. (2022). Deep learning in neural networks: An overview. Neural Networks, 61, 85–117. [Google Scholar] [Crossref]
21. Smith, J., & Agrawal, R. (2022). Encryption methods for secure AI datasets. Journal of Cryptographic Engineering, 12(3), 201–215. [Google Scholar] [Crossref]
22. Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser Ł, Polosukhin I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30, 5998–6008. [Google Scholar] [Crossref]
23. Yang, L., & Lerch, A. (2020). On the evaluation of generative models in music. Transactions of the International Society for Music Information Retrieval, 3(1), 241–256. [Google Scholar] [Crossref]
24. Yin, W., Li, Z., & Sun, M. (2023). Advances in transformer-based architectures for sequential modeling. Artificial Intelligence Review, 56(5), 3211–3235. [Google Scholar] [Crossref]
25. Yi, Y. (2026). Depth-structured music recurrence: Budgeted recurrent attention for full-piece symbolic music modeling. arXiv preprint arXiv:2602.19816. [Google Scholar] [Crossref]
26. Zhang, L., Chen, Y., & Li, H. (2023). Sequential modeling for music generation using LSTM and GRU networks. Expert Systems with Applications, 224, 119875. [Google Scholar] [Crossref]
Metrics
Views & Downloads
Similar Articles
- What the Desert Fathers Teach Data Scientists: Ancient Ascetic Principles for Ethical Machine-Learning Practice
- Comparative Analysis of Some Machine Learning Algorithms for the Classification of Ransomware
- Comparative Performance Analysis of Some Priority Queue Variants in Dijkstra’s Algorithm
- Transfer Learning in Detecting E-Assessment Malpractice from a Proctored Video Recordings.
- Dual-Modal Detection of Parkinson’s Disease: A Clinical Framework and Deep Learning Approach Using NeuroParkNet