Deep Learning and Spectrum Analysis for Timbre Evaluation in Guzheng Performances
Abstract
Guzheng, a representative of Chinese traditional instrumental music, has long relied on subjective timbre evaluation without systematic modeling. This study integrates spectrum analysis and deep learning to construct an automatic sound quality evaluation framework. Audio samples across multiple playing styles were collected and processed into frequency- and time-domain features, including spectral centroid, entropy, and energy density. CNN and SVR models were compared in predicting expert scores. Results show that CNN achieved an MSE of 0.017 (95% CI [0.014, 0.020]) and R² of 0.942, significantly outperforming SVR (p < 0.01). Prediction accuracy reached 91.5% in classical style, with deviations from expert scores within 3.5%. Statistical validation and ANOVA confirmed robustness across styles. These findings demonstrate that spectral structure plays a leading role in timbre perception and that deep networks are effective in modeling complex instrumental signals. The framework provides a quantitative basis for guzheng performance analysis and intelligent teaching feedback, with potential for broader application.References
Zeba G, Dabic M, Cicak M, Daim T, Yalcin H. Technology mining: Artificial intelligence in manufacturing. Technological Forecasting and Social Change. 2021, 171: 120971. doi: 10.1016/j.techfore.2021.120971
Newman J, Mintrom M, O'Neill D. Digital technologies, artificial intelligence, and bureaucratic transformation. Futures. 2022, 136: 102886. doi: 10.1016/j.futures.2021.102886.
Zhai SX, Liu ZP. Artificial intelligence technology innovation and firm productivity: Evidence from China. Finance Research Letters. 2023, 58: 104437. doi: 10.1016/j.frl.2023.104437.
Vannuccini S, Prytkova E. Artificial Intelligence's new clothes? A system technology perspective. Journal of Information Technology. 2024, 39 (2): 317-38. doi: 10.1177/02683962231197824.
Ma DC, Wu WW. Does artificial intelligence drive technology convergence? Evidence from Chinese manufacturing companies. Technology in Society. 2024, 79: 102715. doi: 10.1016/j.techsoc.2024.102715.
Jin XY, Li XW. Painting style-based recognition of potters: using convolutional neural network techniques. Archaeological and Anthropological Sciences. 2025, 17 (5): 100. doi:10.1007/s12520-025-02206-6.
Xu YT, Yamashita A, Uno K, Kawashima T, Amano K. Prediction of alpha power using multiple subjective measures and autonomic responses. Psychophysiology. 2025, 62 (3): e70028. doi:10.1111/psyp.70028.
Tilmatine M, Lüdtke J, Jacobs AM. Predicting subjective ratings of affect and comprehensibility with text features: a reader response study of narrative poetry. Frontiers in Psychology. 2024, 15: 1431764. doi:10.3389/fpsyg.2024.1431764.
Blackwater D, Murtazashvili I, Weiss MBH. Polycentric systems for spectrum management: the case of Indigenous and tribal spectrum sovereignty. European Journal of Law and Economics. 2024, 57 (3): 465–491. doi:10.1007/s10657-024-09803-1.
Vawda MI, Lottering R, Mutanga O, Peerbhay K, Sibanda M. Comparing the utility of artificial neural networks (ANN) and convolutional neural networks (CNN) on Sentinel-2 MSI to estimate dry season aboveground grass biomass. Sustainability. 2024, 16 (3): 1051. doi:10.3390/su16031051.
Nasab RZ, Mohseni H, Montazeri M, Ghasemian F, Amin S. AFEX-Net: Adaptive feature extraction convolutional neural network for classifying computerized tomography images. Digital Health. 2024, 10: 470-473. 10:20552076241232882. doi:10.1177/20552076241232882.
Lu ZH, Wang JL, Wang FQ, Wu ZM. Application of graph frequency attention convolutional neural networks in depression treatment response. Front Psychiatry. 2023, 14: 1244208. doi:10.3389/fpsyt.2023.1244208.
Corlazzoli G, Desender K, Gevers W. Feeling and deciding: subjective experiences rather than objective factors drive the decision to invest cognitive control. Cognition. 2023, 240:105587. doi:10.1016/j.cognition.2023.105587.
Farahi SMMM, Leth-Steensen C. Latent profile analysis of autism spectrum quotient. Current Psychology. 2023, 42 (34): 30029–30036. doi:10.1007/s12144-022-03990-3.
Zou F, Li XY, Chen FF, Wang Y, Wang L, Wang YF, et al. P2 manifests subjective evaluation of reward processing under social comparison. Frontiers in Psychology. 2022; 13: 817529. doi:10.3389/fpsyg.2022.817529.
Ross CA, Litvin J, Ryals A, Kaminski PL. The autonomic spectrum questionnaire: a factor analysis. Current Psychology. 2023, 42 (5): 4264–4271. doi:10.1007/s12144-021-01789-2.
Zhang YD, Satapathy SC, Guttery DS, Górriz JM, Wang SH. Improved breast cancer classification through combining graph convolutional network and convolutional neural network. Information Processing & Management. 2021, 58 (2): 102439. doi:10.1016/j.ipm.2020.102439.
Alilovic J, Slagter HA, van Gaal S. Subjective visibility report is facilitated by conscious predictions only. Consciousness and Cognition 2021, 87: 103048. doi:10.1016/j.concog.2020.103048.
Cui H, Yuan GG, Liu N, Xu MY, Song HS. Convolutional neural network for recognizing highway traffic congestion. Journal of Intelligent Transportation Systems. 2020, 24 (3): 279–289. doi:10.1080/15472450.2020.1742121.
Lodi-Smith J, Rodgers JD, Cunningham SA, Lopata C, Thomeer ML. Meta-analysis of Big Five personality traits in autism spectrum disorder. Autism Research. 2019, 23 (3): 556–565. doi:10.1177/1362361318766571.
Vile JL, Gillard JW, Harper PR, Knight VA. Predicting ambulance demand using singular spectrum analysis. Journal of the Operational Research Society. 2012, 63 (11): 1556–1565. doi:10.1057/jors.2011.160.
De Zeng. Artificial intelligence choreography for music-driven dance generation using multilayer perceptron model. Informatica. 2025;49(20): 137-148. https://doi.org/10.31449/inf.v49i20.8103
Xu Q. Adaptive semantic perception model for image processing and pattern recognition based on deep learning. Informatica. 2025;49(29):17-34. https://doi.org/10.31449/inf.v49i29.8724
Shijina V, Unni A, John SJ. Similarity measure of multiple sets and its application to pattern recognition. Informatica. 2020;44(3): 335-347. https://doi.org/10.31449/inf.v44i3.2872
DOI:
https://doi.org/10.31449/inf.v49i35.10133Downloads
Published
How to Cite
Issue
Section
License
I assign to Informatica, An International Journal of Computing and Informatics ("Journal") the copyright in the manuscript identified above and any additional material (figures, tables, illustrations, software or other information intended for publication) submitted as part of or as a supplement to the manuscript ("Paper") in all forms and media throughout the world, in all languages, for the full term of copyright, effective when and if the article is accepted for publication. This transfer includes the right to reproduce and/or to distribute the Paper to other journals or digital libraries in electronic and online forms and systems.
I understand that I retain the rights to use the pre-prints, off-prints, accepted manuscript and published journal Paper for personal use, scholarly purposes and internal institutional use.
In certain cases, I can ask for retaining the publishing rights of the Paper. The Journal can permit or deny the request for publishing rights, to which I fully agree.
I declare that the submitted Paper is original, has been written by the stated authors and has not been published elsewhere nor is currently being considered for publication by any other journal and will not be submitted for such review while under review by this Journal. The Paper contains no material that violates proprietary rights of any other person or entity. I have obtained written permission from copyright owners for any excerpts from copyrighted works that are included and have credited the sources in my article. I have informed the co-author(s) of the terms of this publishing agreement.
Copyright © Slovenian Society Informatika
