Data‑Driven Analysis of Musical Form and Harmonic Structure in AI‑Generated Popular Music: A Case Study with Suno and Udio

David Dalmazzo; Laura Cros Vila; Luca Casini; Bob L.T. Sturm

doi:10.5334/tismir.348

Data‑Driven Analysis of Musical Form and Harmonic Structure in AI‑Generated Popular Music: A Case Study with Suno and Udio

Transactions of the International Society for Music Information Retrieval

Volume 9 (2026): Issue 1

By: David Dalmazzo , Laura Cros Vila , Luca Casini and Bob L.T. Sturm

Open Access

|Sep 2026

Abstract

We present a large‑scale analysis of the harmonic and formal structures of music generated by users of two AI music‑generation platforms, Suno and Udio, compared with human‑composed popular music. We address three questions: (1) What considerations arise when applying music information retrieval tools to AI‑generated music? (2) How does AI‑generated music’s harmony relate to human‑composed popular music? (3) What structural similarities or differences exist? We analyze 20,002 Suno tracks, 20,002 Udio tracks, and 20,010 human‑composed recordings, performing harmonic analysis with a state‑of‑the‑art chord‑estimation model and structural analysis with a joint beat‑tracking and semantic section‑labeling model. We model harmonic progression and musical form using trigram and tetragram analyses of detected chords and labeled sections, quantifying their diversity across the three collections. Music in our Suno collection concentrates more strongly on Western pop harmonic patterns, such as I‑V‑vi‑IV, than music in either our Udio or human collection, and tends toward simple verse‑chorus repetitions, whereas Udio exhibits wider structural variety. These findings suggest that music generated by Suno (up to version 3.5) sits in a narrower harmonic and formal space than that of Udio and human‑composed music. Fully automated, open‑source music information retrieval tools now support corpus‑scale harmonic and structural analysis, although chord estimation remains an open challenge, particularly for inversions, extensions, and substituted dominants. We release our beat‑level chord annotations, metadata, section boundaries, self‑similarity matrices, audio links, and code under an MIT license.

References

Agostinelli, A., Denk, T. I., Borsos, Z., Engel, J., Verzetti, M., Caillon, A., Huang, Q., Jansen, A., Roberts, A., Tagliasacchi, M., Sharifi, M., Zeghidour, N., and Frank, C. (2023). Musiclm: Generating music from text. arXiv preprint arXiv:2301.11325. 10.48550/arXiv.2301.11325.
Open DOI Search in Google Scholar Back to article
Ariza, C. (2009). The interrogator as critic: The Turing test and the evaluation of generative music systems. Computer Music Journal, 33(2), 48–70. 10.1162/comj.2009.33.2.48.
Open DOI Search in Google Scholar Back to article
Benjamini, Y., and Hochberg, Y. (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B (Methodological), 57(1), 289–300. 10.1111/j.2517-6161.1995.tb02031.x.
Open DOI Search in Google Scholar Back to article
Bertin‑Mahieux, T., Ellis, D. P., Whitman, B., and Lamere, P. (2011). The million song dataset. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 591–596. 10.5281/zenodo.1415820.
Open DOI Search in Google Scholar Back to article
Biamonte, N. (2010). Triadic modal and pentatonic patterns in rock music. Music Theory Spectrum, 32(2), 95–110. 10.1525/mts.2010.32.2.95.
Open DOI Search in Google Scholar Back to article
Böck, S., Korzeniowski, F., Schlüter, J., Krebs, F., and Widmer, G. (2016). Madmom: A new Python audio and music signal processing library. In Proceedings of the 24th ACM International Conference on Multimedia (ACM‑MM), pp. 1174–1178. ACM. 10.1145/2964284.2973795.
Open DOI Search in Google Scholar Back to article
Bommasani, R., Creel, K. A., Kumar, A., Jurafsky, D., and Liang, P. S. (2022). Picking on the same person: Does algorithmic monoculture lead to outcome homogenization? Advances in Neural Information Processing Systems, 35, 3663–3678. 10.52202/068431-0265.
Open DOI Search in Google Scholar Back to article
Buisson, M., McFee, B., and Essid, S. (2024). Using pairwise link prediction and graph attention networks for music structure analysis. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 207–214. 10.5281/zenodo.14877311.
Open DOI Search in Google Scholar Back to article
Buisson, M., McFee, B., Essid, S., and Crayencour, H.‑C. (2022). Learning multi‑level representations for hierarchical music structure analysis. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 591–597. 10.5281/zenodo.7343060.
Open DOI Search in Google Scholar Back to article
Burgoyne, J. A., Wild, J., and Fujinaga, I. (2011). An expert ground truth set for audio chord recognition and music analysis. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), Vol. 11, pp. 633–638. 10.5281/zenodo.1417547.
Open DOI Search in Google Scholar Back to article
Casini, L., Cros Vila, L., Dalmazzo, D., Kaila, A.‑K., and Sturm, B. L. T. (2026). Data‑driven analysis of text‑conditioning in AI‑generated music: A case study with Suno and Udio. Transactions of the International Society for Music Information Retrieval, 9(1), 194–209. 10.5334/tismir.273.
Open DOI Search in Google Scholar Back to article
Collins, N. (2025). Recording artist career comparison through audio content analysis. Royal Society Open Science, 12(7), 241647. 10.1098/rsos.241647.
Open DOI Search in Google Scholar Back to article
Copet, J., Kreuk, F., Gat, I., Remez, T., Kant, D., Synnaeve, G., Adi, Y., and Défossez, A. (2023). Simple and controllable music generation. Advances in Neural Information Processing Systems, 36, 47704–47720. 10.52202/075280-2066.
Open DOI Search in Google Scholar Back to article
Cros Vila, L., Sturm, B., Casini, L., and Dalmazzo, D. (2025). The AI music arms race: On the detection of AI‑generated music. Transactions of the International Society for Music Information Retrieval, 8(1), 179–194. 10.5334/tismir.254.
Open DOI Search in Google Scholar Back to article
Cuthbert, M. S., and Ariza, C. (2010). music21: A toolkit for computer‑aided musicology and symbolic music data. In Proceedings of the 11th International Society for Music Information Retrieval Conference (ISMIR), pp. 637–642. 10.5281/zenodo.1416114.
Open DOI Search in Google Scholar Back to article
Déguernel, K., and Sturm, B. L. T. (2023). Bias in favour or against computational creativity: A survey and reflection on the importance of socio‑cultural context in its evaluation. In Proceedings of the International Conference on Computational Creativity (ICCC).
Search in Google Scholar Back to article
Dervakos, E., Filandrianos, G., and Stamou, G. (2021). Heuristics for evaluation of AI generated music. In 2020 25th International Conference on Pattern Recognition (ICPR), pp. 9164–9171. IEEE. 10.1109/ICPR48806.2021.9413310.
Open DOI Search in Google Scholar Back to article
European Union. (2019, May 17). Directive (EU) 2019/790 of the European Parliament and of the Council of 17 April 2019 on copyright and related rights in the Digital Single Market. Official Journal of the European Union, L 130, 92–125. http://data.europa.eu/eli/dir/2019/790/oj.
Search in Google Scholar Back to article
Evans, Z., Parker, J. D., Carr, C., Zukowski, Z., Taylor, J., and Pons, J. (2025). Stable audio open. In ICASSP 2025–2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1–5. IEEE. 10.1109/ICASSP49660.2025.10888461.
Open DOI Search in Google Scholar Back to article
Foote, J. (2000). Automatic audio segmentation using a measure of audio novelty. In 2000 IEEE International Conference on Multimedia and Expo (ICME 2000) Proceedings. Latest Advances in the Fast Changing World of Multimedia (Cat. No.00TH8532), Vol. 1, pp. 452–455. IEEE. 10.1109/ICME.2000.869637.
Open DOI Search in Google Scholar Back to article
Humphrey, E. J., and Bello, J. P. (2015). Four timely insights on automatic chord estimation. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), Vol. 10, pp. 673–679. 10.5281/zenodo.1417549.
Open DOI Search in Google Scholar Back to article
Jiang, J., Chen, K., Li, W., and Xia, G. (2019). Large‑vocabulary chord transcription via chord structure decomposition. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 644–651. 10.5281/zenodo.3527892.
Open DOI Search in Google Scholar Back to article
Kim, T., and Nam, J. (2023). All‑in‑one metrical and functional structure analysis with neighborhood attentions on demixed audio. In IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA). 10.1109/WASPAA58266.2023.10248148.
Open DOI Search in Google Scholar Back to article
Levy, M., and Sandler, M. (2008). Structural segmentation of musical audio by constrained clustering. IEEE Transactions on Audio, Speech, and Language Processing, 16(2), 318–326. 10.1109/TASL.2007.910781.
Open DOI Search in Google Scholar Back to article
Lin, Z., Ehsan, U., Agarwal, R., Dani, S., Vashishth, V., and Riedl, M. (2023). Beyond prompts: Exploring the design space of mixed‑initiative co‑creativity systems. In Proceedings of the 14th International Conference on Computational Creativity (ICCC), pp. 64–73. https://computationalcreativity.net/iccc23/.
Search in Google Scholar Back to article
Marmoret, A., Cohen, J. E., and Bimbot, F. (2023). Barwise music structure analysis with the correlation block‑matching segmentation algorithm. Transactions of the International Society for Music Information Retrieval, 6(1), 167–185. 10.5334/tismir.167.
Open DOI Search in Google Scholar Back to article
Mauch, M., and Dixon, S. (2009). Simultaneous estimation of chords and musical context from audio. IEEE Transactions on Audio, Speech, and Language Processing, 18(6), 1280–1289. 10.1109/TASL.2009.2032947.
Open DOI Search in Google Scholar Back to article
Mauch, M., and Dixon, S. (2010). Approximate note transcription for the improved identification of difficult chords. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 135–140. 10.5281/zenodo.1416598.
Open DOI Search in Google Scholar Back to article
Mauch, M., MacCallum, R. M., Levy, M., and Leroi, A. M. (2015). The evolution of popular music: USA 1960–2010. Royal Society Open Science, 2(5), 150081. 10.1098/rsos.150081.
Open DOI Search in Google Scholar Back to article
McFee, B., and Bello, J. P. (2017). Structured training for large‑vocabulary chord recognition. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 188–194. 10.5281/zenodo.1414880.
Open DOI Search in Google Scholar Back to article
McFee, B., and Ellis, D. P. (2014). Learning to segment songs with ordinal linear discriminant analysis. In 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 5197–5201. IEEE. 10.1109/ICASSP.2014.6854594.
Open DOI Search in Google Scholar Back to article
McFee, B., Raffel, C., Liang, D., Ellis, D. P., McVicar, M., Battenberg, E., and Nieto, O. (2025). librosa. 10.5281/zenodo.15006942.
Open DOI Search in Google Scholar Back to article
Müller, M. (2021). Fundamentals of Music Processing: Using Python and Jupyter Notebooks. Springer. 10.1007/978-3-030-69808-9.
Open DOI Search in Google Scholar Back to article
Ni, Y., McVicar, M., Santos‑Rodriguez, R., and De Bie, T. (2012). An end‑to‑end machine learning system for harmonic analysis of music. IEEE Transactions on Audio, Speech, and Language Processing, 20(6), 1771–1783. 10.1109/TASL.2012.2188516.
Open DOI Search in Google Scholar Back to article
Nieto, O., Mysore, G. J., Wang, C.‑I., Smith, J. B., Schlüter, J., Grill, T., and McFee, B. (2020). Audio‑based music structure analysis: Current trends, open challenges, and applications. Transactions of the International Society for Music Information Retrieval, 3(1). 10.5334/tismir.54.
Open DOI Search in Google Scholar Back to article
Odekerken, D., Koops, H. V., and Volk, A. (2021). Improving audio chord estimation by alignment and integration of crowd‑sourced symbolic music. Transactions of the International Society for Music Information Retrieval, 4(1), 141–155. 10.5334/tismir.81.
Open DOI Search in Google Scholar Back to article
Oramas, S., Barbieri, F., Nieto, O., and Serrà, X. (2018). Multimodal deep learning for music genre classification. Transactions of the International Society for Music Information Retrieval, 1(1), 4–22. 10.5334/tismir.10.
Open DOI Search in Google Scholar Back to article
Park, J., Choi, K., Jeon, S., Kim, D., and Park, J. (2019). A bi‑directional transformer for musical chord recognition. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 620–627. 10.5281/zenodo.3527886.
Open DOI Search in Google Scholar Back to article
Pasquier, P., Burnett, A., Thomas, N. G., Maxwell, J. B., Eigenfeldt, A., and Loughin, T. (2016). Investigating listener bias against musical metacreativity. In Proceedings of the International Conference on Computational Creativity (ICCC).
Search in Google Scholar Back to article
Pauwels, J., and Peeters, G. (2013). Evaluating automatically estimated chord sequences. In 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 749–753. IEEE. 10.1109/ICASSP.2013.6637748.
Open DOI Search in Google Scholar Back to article
Peeters, G. (2023). Self‑similarity‑based and novelty‑based loss for music structure analysis. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 749–756. 10.5281/zenodo.10265397.
Open DOI Search in Google Scholar Back to article
Pelly, L. (2025). Mood Machine: The Rise of Spotify and the Costs of the Perfect Playlist. Simon and Schuster.
Search in Google Scholar Back to article
Poltronieri, A., Serrà, X., and Rocamora, M. (2025). From discord to harmony: Decomposed consonance‑based training for improved audio chord estimation. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 492–500. 10.5281/zenodo.17811559.
Open DOI Search in Google Scholar Back to article
Poppe, T., Lopes, L., and Figueiredo, F. (2025). The polyphonic audio to roman corpus. In Proceedings of the 12th International Conference on Digital Libraries for Musicology, pp. 72–80. 10.1145/3748336.3748345.
Open DOI Search in Google Scholar Back to article
Ragot, M., Martin, N., and Cojean, S. (2020). AI‑generated vs. human artworks: A perception bias towards artificial intelligence? In Proceedings of the Conference on Human Factors in Computing Systems (CHI). 10.1145/3334480.3382892.
Open DOI Search in Google Scholar Back to article
Richards, M. (2017). Tonal ambiguity in popular music’s axis progressions. Music Theory Online, 23(3). 10.30535/mto.23.3.6.
Open DOI Search in Google Scholar Back to article
Rouard, S., Massa, F., and Défossez, A. (2023). Hybrid transformers for music source separation. In ICASSP 2023–2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1–5. IEEE. 10.1109/ICASSP49357.2023.10096956.
Open DOI Search in Google Scholar Back to article
Salamon, J., Nieto, O., and Bryan, N. J. (2021). Deep embeddings and section fusion improve music segmentation. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 594–601. 10.5281/zenodo.5624371.
Open DOI Search in Google Scholar Back to article
Serrà, J., Corral, Á., Boguñá, M., Haro, M., and Arcos, J. L. (2012a). Measuring the evolution of contemporary western popular music. Scientific Reports, 2(1), 1–6. 10.1038/srep00521.
Open DOI Search in Google Scholar Back to article
Serrà, J., Müller, M., Grosche, P., and Arcos, J. L. (2012b). Unsupervised detection of music boundaries by time series structure features. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 26, pp. 1613–1619. 10.1609/aaai.v26i1.8328.
Open DOI Search in Google Scholar Back to article
Shank, D. B., Stefanik, C., Stuhlsatz, C., Kacirek, K., and Belfi, A. M. (2022). AI composer bias: Listeners like music less when they think it was composed by an AI. Journal of Experimental Psychology: Applied. 10.1037/xap0000447.
Open DOI Search in Google Scholar Back to article
Sturm, B. L., and Ben‑Tal, O. (2017). Taking the models back to music practice: Evaluating generative transcription models built using deep learning. Journal of Creative Music Systems, 2(1). 10.5920/jcms.2017.09.
Open DOI Search in Google Scholar Back to article
Sturm, B. L. T., Déguernel, K., Huang, R. S., Kaila, A.‑K., Jääskeläinen, P., Kanhov, E., Cros Vila, L., Dalmazzo, D., Casini, L., Bown, O. R., Collins, N., Drott, E., Sterne, J., Holzapfel, A., and Ben‑Tal, O. (2024). AI music studies: Preparing for the coming flood. In AIMC 2024, Oxford, United Kingdom. 10.5281/zenodo.15110181.
Open DOI Search in Google Scholar Back to article
Tan, S. (2024). Are we all musicians now? Authenticity, musicianship, and AI music generator Suno. 10.31235/osf.io/4nt8z.
Open DOI Search in Google Scholar Back to article
Temperley, D., and Clercq, T. d. (2013). Statistical analysis of harmony and melody in rock music. Journal of New Music Research, 42(3), 187–204. 10.1080/09298215.2013.788039.
Open DOI Search in Google Scholar Back to article
Wang, J.‑C., Hung, Y.‑N., and Smith, J. B. (2022). To catch a chorus, verse, intro, or anything else: Analyzing a song with structural functions. In ICASSP 2022–2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 416–420. IEEE. 10.1109/ICASSP43922.2022.9747252.
Open DOI Search in Google Scholar Back to article
Waseem Akram, M., Dettori, S., Colla, V., and Carlo Buttazzo, G. (2025). Chordformer: A conformer‑based architecture for large‑vocabulary audio chord recognition. IEEE Transactions on Audio, Speech, and Language Processing. 10.1109/TASLPRO.2025.3646468.
Open DOI Search in Google Scholar Back to article
Wiggins, J. (2007). Compositional process in music. In International Handbook of Research in Arts Education, pp. 453–476. Springer. 10.1007/978-1-4020-3052-9_29.
Open DOI Search in Google Scholar Back to article
Wiggins, J. (2016). Musical agency. In G. E. McPherson (Ed.), The Child as Musician: A Handbook of Musical Development, pp. 102–121. Oxford University Press. 10.1093/acprof:oso/9780198744443.003.0006.
Open DOI Search in Google Scholar Back to article
Wu, S.‑L., and Yang, Y.‑H. (2020). The jazz transformer on the front line: Exploring the shortcomings of AI‑composed music through quantitative measures. In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), pp. 142–149. 10.5281/zenodo.4245390.
Open DOI Search in Google Scholar Back to article
Zlatkov, D., Ens, J., and Pasquier, P. (2023). Searching for human bias against AI‑composed music. In Proceedings of the International Conference on Artificial Intelligence in Music, Sound, Art and Design (EvoMUSART). 10.1007/978-3-031-29956-8_20.
Open DOI Search in Google Scholar Back to article

Articles in this issue

DOI: https://doi.org/10.5334/tismir.348 | Journal eISSN: 2514-3298

Journal RSS Feed

Language: English

Page range: 491 - 509

Submitted on: Oct 8, 2025

Accepted on: Aug 6, 2026

Published on: Sep 10, 2026

Published by: Ubiquity Press

In partnership with: Paradigm Publishing Services

Keywords:

automatic chord estimation,

exploratory data analysis

© 2026 David Dalmazzo, Laura Cros Vila, Luca Casini, Bob L.T. Sturm, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.

Volume 9 (2026): Issue 1