COMPARATIVE EVALUATION OF LSTM VARIANTS FOR TRADITIONAL MUSICAL INSTRUMENT AUDIO CLASSIFICATION USING MFCC FEATURES

Ridwan Andi Kambau; Muhammad Syawal Idil Fitrah Baharuddin

doi:10.24252/instek.v11i1.66225

Authors

Ridwan Andi Kambau UIN Alauddin Makassar
Muhammad Syawal Idil Fitrah Baharuddin Universitas Hasanuddin

DOI:

https://doi.org/10.24252/instek.v11i1.66225

Keywords:

Audio Classification, Attention-Based LSTM, MFCC Features, Traditional Musical Instruments, Deep Learning

Abstract

Classifying traditional musical instrument audio remains challenging due to limited labeled data, strong acoustic variability, and spectral similarity across instruments. This paper proposes an attention-based Long Short-Term Memory (LSTM) model for traditional instrument sound classification using Mel-Frequency Cepstral Coefficients (MFCC) as the feature representation. Three LSTM variants, Bidirectional LSTM, Residual LSTM, and Attention-based LSTM are investigated to identify the most effective temporal architecture for this task. The attention mechanism is specifically integrated to enable the model to prioritize discriminative temporal segments, such as unique attack phases and harmonic decay, which are often obscured in traditional instruments. The dataset comprises 1,000 audio samples from 10 traditional instrument classes. All samples are normalized to 3-second duration and augmented via pitch shifting, time stretching, and additive noise to improve generalization. Using 5-fold cross-validation, the Attention-based LSTM consistently achieves the highest performance, with average accuracy 96.73%. This superiority stems from the mechanism’s ability to surpress irrelevant noise frames while focusing on key spectral-temporal features. Robustnes experiments maintain accuracy above 90% under noisy conditions, suggesting that coupling MFCC with attention-enchanced modeling provides a robust solution for cultural heritage preservation through digital audio recognition.

Downloads

Download data is not yet available.

References

[1] Choi, Keunwoo, et al. "A comparison of audio signal preprocessing methods for deep neural networks on music tagging." 2018 26th European signal processing conference (EUSIPCO). IEEE, 2018.

[2] Blaszke M, Kostek B. Musical Instrument Identification Using Deep Learning Approach. Sensors (Basel). 2022 Apr 15;22(8):3033. doi: 10.3390/s22083033. PMID: 35459018; PMCID: PMC9025072.

[3] Han Y, Kim J, Lee K. Deep convolutional neural networks for predominant instrument recognition in polyphonic music. IEEE/ACM Transactions on Audio, Speech, and Language Processing. 2016 Nov 23;25(1):208-21.

[4] B. Logan, “Mel frequency cepstral coefficients for music modeling,” in Proc. 1st Int. Symp. Music Inf. Retrieval (ISMIR), Plymouth, MA, USA, 2000, pp. 1–11. [Online]. Available: http://ismir2000.ismir.net/papers/logan_paper.pdf

[5] J. Pons and X. Serra, "Designing efficient architectures for modeling temporal features with convolutional neural networks," 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, LA, USA, 2017, pp. 2472-2476, doi: 10.1109/ICASSP.2017.7952601.

[6] Sunitha G, Arunachalam R, Abd-Elnaby M, Eid MMA, Rashed ANZ. A comparative analysis of deep neural network architectures for the dynamic diagnosis of COVID-19 based on acoustic cough features. Int J Imaging Syst Technol. 2022 Sep;32(5):1433-1446. doi: 10.1002/ima.22749. Epub 2022 May 21. PMID: 35941929; PMCID: PMC9348187.

[7] Khalid, Syed Sibtain. "Attention-Enhanced Deep Learning for Urban Environmental Sound Classification." Anusandhanvallari (2024): 2735-2743.

[8] Demšar, Janez. "Statistical comparisons of classifiers over multiple data sets." Journal of Machine learning research 7.Jan (2006): 1-30.

[9] Bittner RM, McFee B, Salamon J, Li P, Bello JP. Deep Salience Representations for F0 Estimation in Polyphonic Music. InISMIR 2017 Oct 23 (pp. 63-70).

[10] Ko, T., Peddinti, V., Povey, D., & Khudanpur, S. (2015, September). Audio augmentation for speech recognition. In Interspeech (Vol. 2015, p. 3586).

[11] Sharan RV, Xiong H, Berkovsky S. Benchmarking Audio Signal Representation Techniques for Classification with Convolutional Neural Networks. Sensors (Basel). 2021 May 14;21(10):3434. doi: 10.3390/s21103434. PMID: 34069189; PMCID: PMC8156023.

[12] Schuster, Mike, and Kuldip K. Paliwal. "Bidirectional recurrent neural networks." IEEE transactions on Signal Processing 45.11 (1997): 2673-2681.

[13] Das S, Tariq A, Santos T, et al. Recurrent Neural Networks (RNNs): Architectures, Training Tricks, and Introduction to Influential Research. 2023 Jul 23. In: Colliot O, editor. Machine Learning for Brain Disorders [Internet]. New York, NY: Humana; 2023. Chapter 4. Available from: https://www.ncbi.nlm.nih.gov/books/NBK597502/ doi: 10.1007/978-1-0716-3195-9_4

[14] Z. Lin et al., “A structured self-attentive sentence embedding,” arXiv preprint arXiv:1703.03130, 2017. [Online]. Available: https://arxiv.org/abs/1703.03130

[15] D. M. W. Powers, “Evaluation: From precision, recall and F-measure to ROC, informedness, markedness and correlation,” arXiv preprint arXiv:2010.16061, 2020. [Online]. Available: https://arxiv.org/abs/2010.16061

[16] Mienye, Ibomoiye Domor, Theo G. Swart, and George Obaido. "Recurrent neural networks: A comprehensive review of architectures, variants, and applications." Information 15.9 (2024): 517.

[17] J. Pons, O. Slizovskaia, R. Gong, E. Gómez, and X. Serra, “Timbre analysis of music audio signals with convolutional neural networks,” arXiv preprint arXiv:1703.06697, 2017. [Online]. Available: https://arxiv.org/abs/1703.06697.

[18] K. Choi, G. Fazekas, K. Cho, and M. Sandler, “A tutorial on deep learning for music information retrieval,” arXiv preprint arXiv:1709.04396, 2018. [Online]. Available: https://arxiv.org/abs/1709.04396

[19] S. Hershey et al., “CNN architectures for large-scale audio classification,” arXiv preprint arXiv:1609.09430, 2017. [Online]. Available: https://arxiv.org/abs/1609.09430

[20] B. McFee et al., "librosa: Audio and music signal analysis in Python," in Proc. 14th Python Sci. Conf., vol. 8, 2015, pp. 18–25.

[21] J. Gemmeke et al., “Audio Set: An ontology and human-labeled dataset for audio events,” in Proc. IEEE Int. Conf. Acoust. Speech Signal Process. (ICASSP), 2017, pp. 776–780, doi: 10.1109/ICASSP.2017.7952261.