Abstract
This paper compares two tokenization strategies for modeling chord progressions using the encoder transformer architecture trained with a large dataset of chord progressions in a variety of styles. The first strategy includes a tokenization method treating all different chords as unique elements, which results in a vocabulary of 5202 independent tokens. The second strategy expresses the chords as a dynamic tuple describing root, nature (e.g., major, minor, diminished, etc.), and extensions (e.g., additions or alterations), producing a specific vocabulary of 59 tokens related to chords and 75 tokens for style, bars, form, and format. In the second approach, MIDI embeddings are added into the positional embedding layer of the transformer architecture, with an array of eight values related to the notes forming the chords. We propose a trigram analysis addition to the dataset to compare the generated chord progressions with the training dataset, which reveals common progressions and the extent to which a sequence is duplicated. We analyze progressions generated by the models comparing HITS@k metrics and human evaluation of 10 participants, rating the plausibility of the progressions as potential music compositions from a musical perspective. The second model reported lower validation loss, better metrics, and more musical consistency in the suggested progressions.
This paper is an outcome of: MUSAiC, a project that has received funding from the European Research Council under the European Union’s Horizon 2020 research and innovation program (Grant agreement No. 864189); and a Margarita Salas Grant, UPF, Barcelona, Spain.
References
Agostinelli, A., et al.: Musiclm: generating music from text. arXiv preprint arXiv:2301.11325 (2023)
Ali, M., et al.: Bringing light into the dark: a large-scale evaluation of knowledge graph embedding models under a unified framework. IEEE Trans. Pattern Anal. Mach. Intell. 44(12), 8825–8845 (2021)
Briot, J.P., Hadjeres, G., Pachet, F.: Deep Learning Techniques for Music Generation. Springer, Cham (2019). https://doi.org/10.1007/978-3-319-70163-9
Caillon, A., Esling, P.: Rave: a variational autoencoder for fast and high-quality neural audio synthesis. ArXiv e-prints (2021)
Chen, T.P., Su, L.: Attend to chords: improving harmonic analysis of symbolic music using transformer-based models. Trans. Int. Soc. Music Inf. Retrieval 4(1) (2021)
Choi, K., Fazekas, G., Sandler, M.: Text-based LSTM networks for automatic music composition. In: Proceedings of 1st Conference on Computer Simulation of Musical Creativity. Huddersfield, UK (2016)
Huang, C.Z.A., Duvenaud, D., Gajos, K.Z.: Chordripple: recommending chords to help novice composers go beyond the ordinary. In: Proceedings of the 21st International Conference on Intelligent User Interfaces, pp. 241–250 (2016)
Huang, C.Z.A., et al.: Music transformer: generating music with long-term structure (2018). arXiv preprint arXiv:1809.04281 (2018)
Li, S., Sung, Y.: MRBERT: pre-training of melody and rhythm for automatic music generation. Mathematics 11(4), 798 (2023)
Li, S., Sung, Y.: Transformer-based seq2seq model for chord progression generation. Mathematics 11(5), 1111 (2023)
Pfleiderer, M., Frieler, K., Abeßer, J., Zaddach, W.G., Burkhart, B. (eds.): Inside the Jazzomat - New Perspectives for Jazz Research. Schott Campus (2017)
van den Oord, A., et al.: WaveNet: a generative model for raw audio. ArXiv e-prints (1609.03499) (2016)
Vaswani, A., et al.: Attention is all you need. In: Advances in Neural Information Processing Systems, vol. 30 (2017)
Wu, S.L., Yang, Y.H.: The jazz transformer on the front line: exploring the shortcomings of ai-composed music through quantitative measures. arXiv preprint arXiv:2008.01307 (2020)
Zeng, M., Tan, X., Wang, R., Ju, Z., Qin, T., Liu, T.Y.: Musicbert: symbolic music understanding with large-scale pre-training. arXiv preprint arXiv:2106.05630 (2021)
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Dalmazzo, D., Déguernel, K., Sturm, B.L.T. (2024). The Chordinator: Modeling Music Harmony by Implementing Transformer Networks and Token Strategies. In: Johnson, C., Rebelo, S.M., Santos, I. (eds) Artificial Intelligence in Music, Sound, Art and Design. EvoMUSART 2024. Lecture Notes in Computer Science, vol 14633. Springer, Cham. https://doi.org/10.1007/978-3-031-56992-0_4
Download citation
DOI: https://doi.org/10.1007/978-3-031-56992-0_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-56991-3
Online ISBN: 978-3-031-56992-0
eBook Packages: Computer ScienceComputer Science (R0)