Reduced Convolutional Recurrent Neural Network Using MFCC for Music Genre Classification on the GTZAN Dataset
Abstract
This study presents a reduced Convolutional Recurrent Neural Network (CRNN) model for music genre classification, leveraging the GTZAN dataset and Mel-Frequency Cepstral Coefficient (MFCC) feature extraction. Unlike more complex architectures, this model simplifies the CRNN structure to three convolutional layers and two BiLSTM layers, maintaining competitive performance while reducing computational complexity. Key experimental parameters included learning rate tuning (0.1, 0.01, 0.001, and 0.0001) and dropout usage (30% before the BiLSTM layers) to mitigate overfitting. The best configuration, utilizing a learning rate of 0.001 and dropout, achieved an accuracy of 88.64%, outperforming more complex CRNN models by approximately 15%. These results underscore the potential of streamlined architectures in music information retrieval tasks, particularly for applications where computational resources are constrained. Future work will address overfitting issues and refine the dataset for enhanced model performance.References
D. Stefani and L. Turchet, “On the Challenges of Embedded Real-time Music Information Retrieval,” in Proceedings of the International Conference on Digital Audio Effects, DAFx, 2022, pp. 177–184.
P. Ghosh, S. Mahapatra, S. Jana, and R. Kr. Jha, “A Study on Music Genre Classification using Machine Learning,” Int. J. Eng. Bus. Soc. Sci., vol. 1, no. 04, pp. 308–320, 2023, doi: 10.58451/ijebss.v1i04.55.
J. Sang, S. Park, and J. Lee, “Convolutional recurrent neural networks for urban sound classification using raw waveforms,” Eur. Signal Process. Conf., vol. 2018-Septe, pp. 2444–2448, 2018, doi: 10.23919/EUSIPCO.2018.8553247.
V. Bella and S. A. Sanjaya, “Refining Baby Cry Classification using Data Augmentation (TimeStretching and Pitch-Shifting), MFCC Feature Extraction, and LSTM Modeling,” in 2023 7th International Conference on New Media Studies (CONMEDIA), IEEE, 2023. doi: https://doi.org/10.1109/CONMEDIA60526.2023. 10428158.
M. Ashraf et al., “A Hybrid CNN and RNN Variant Model for Music Classification,” Appl. Sci., vol. 13, no. 3, 2023, doi: 10.3390/app13031476.
J. Mendes, “Deep Learning Techniques for Music Genre Classification and Building a Music Recommendation System,” National College of Ireland, 2020.
X. Luo, “Automatic Music Genre Classification based on CNN and LSTM,” Highlights Sci. Eng. Technol., vol. 39, pp. 61–66, 2023, doi: 10.54097/hset.v39i.6494.
M. K. Kumar, K. Sujanasri, B. Neha, G. Akshara, P. Chugh, and P. Haindavi, “Automated Music Genre Classification through Deep Learning Techniques,” E3S Web Conf., vol. 430, 2023, doi: 10.1051/e3sconf/202343001033.
G. Tzanetakis and P. Cook, “Musical genre classification of audio signals,” IEEE Trans. Speech Audio Process., vol. 10, no. 5, pp. 293– 302, 2002, doi: 10.1109/TSA.2002.800560.
G. Tzanetakis and P. Cook, “MARSYAS: A framework for audio analysis”.
S. Y. Yehezkiel and Y. Suyanto, “Music Genre Identification Using SVM and MFCC Feature Extraction,” IJEIS (Indonesian J. Electron. Instrum. Syst., vol. 12, no. 2, p. 115, 2022, doi: 10.22146/ijeis.70898.
I. D. G. Y. A. Wibawa and I. D. M. B. A. Darmawan, “Implementation of audio recognition using mel frequency cepstrum coefficient and dynamic time warping in wirama praharsini,” J. Phys. Conf. Ser., vol. 1722, no. 1, 2021, doi: 10.1088/1742-6596/1722/1/012014.
H. Heriyanto, T. Wahyuningrum, and G. F. Fitriana, “Classification of Javanese Script Hanacara Voice Using Mel Frequency Cepstral Coefficient MFCC and Selection of Dominant Weight Features,” J. Infotel, vol. 13, no. 2, pp. 84– 93, 2021, doi: 10.20895/infotel.v13i2.657.
B. McFee et al., “librosa: Audio and Music Signal Analysis in Python,” Proc. 14th Python Sci. Conf., no. August 2020, pp. 18–24, 2015, doi: 10.25080/majora-7b98e3ed-003.
P. A. Aritonang, M. E. Johan, and I. Prasetiawan, “Aspect-Based Sentiment Analysis on Application Review using CNN (Case Study : Peduli Lindungi Application),” Ultim. Infosys J. Ilmu Sist. Inf., vol. 13, no. 1, pp. 54–61, 2022.
R. Yamashita, M. Nishio, R. Kinh, G. Do, and K. Togashi, “Convolutional neural networks : an overview and application in radiology,” pp. 611– 629, 2018.
N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, “Dropout: A simple way to prevent neural networks from overfitting,” J. Mach. Learn. Res., vol. 15, pp. 1929–1958, 2014.
A. Katrompas and V. Metsis, “Enhancing LSTM Models with Self-attention and Stateful Training,” in Lecture Notes in Networks and Systems, 2022, pp. 217–235. doi: 10.1007/978-3-030-82193- 7_14.
DOI:
https://doi.org/10.31449/inf.v49i17.6885Downloads
Published
How to Cite
Issue
Section
License
Authors retain copyright in their work. By submitting to and publishing with Informatica, authors grant the publisher (Slovene Society Informatika) the non-exclusive right to publish, reproduce, and distribute the article and to identify itself as the original publisher.
All articles are published under the Creative Commons Attribution license CC BY 3.0. Under this license, others may share and adapt the work for any purpose, provided appropriate credit is given and changes (if any) are indicated.
Authors may deposit and share the submitted version, accepted manuscript, and published version, provided the original publication in Informatica is properly cited.
