Automatic classification of instruments from supervised methods of machine learning
Resumo
Sorting instruments is not an easy task for humans or computers, especially when it comes to elements with the same acoustic properties, such as wind, percussion, or strings. Nevertheless, the use of audio descriptors and artificial intelligence techniques can make this duty more accessible. In this paper, three supervised methods, Naive Bayes, decision tree and Support Vector Classifier (SVC) are used to categorize acoustic guitar and bass sounds in a database, using as a parameter the information extracted from audio descriptors. The research resulted in a performance comparison of these three algorithms, considering their hit rates and processing time when classifying samples in different parts of the dataset. After all, some relevant considerations about the feasibility of automatically classifying instruments are presented.
Palavras-chave:
Artificial Intelligence, A-Life and Evolutionary Music Systems, Music Analysis and Synthesis, Music Information Retrieval
Referências
Roberto Piassi Passos Bodo and Marcelo Gomesde Queiroz. Três abordagens para similaridade musical utilizando melodia, ritmo e timbre. Master’s thesis, Universidade Federal de São Paulo, São Paulo, 2018.
Lara Haidar-Ahmad. Music and instrument classification using deep learning technics. In Proceedings of the Conference on Neural Information Processing Systems, pages 1–6. Nips.cc, 2017.
Janet Marques and Pedro Moreno. A study of musical instrument classification using gaussian mixture models andsupport vector machines. Cambridge Research Laboratory-Technical Report Series, 09 1999.
A. Eronen and A. Klapuri. Musical instrument recognition using cepstral coefficients and temporal features. In2000 IEEE International Conference on Acoustics, Speech,and Signal Processing. Proceedings (Cat. No.00CH37100),volume 2, pages II753–II756 vol.2, 2000.
Babak Toghiani-Rizi and Marcus Windmark. Musical instrument recognition using their distinctive characteristicsin artificial neural networks. ArXiv, abs/1705.04971, 2017.
Martin Vetterli Paolo Prandoni. Signal Processing for Communications (Communication and Information Sciences). EPFL Press, 2008.
Ivan Eiji Simurra. A utilização descritores de áudio à análise e composição musical assistidas por computador: um estudo de caso na obra labori ruinae, 2016.
Ivan Eiji Simurra. Análise musical assistida por descritores de Áudio: um estudo de caso da obra reflexões de jônatas manzolli, 2015.
Sound Analysis Pro Editorial. Wiener entropy, Outubro2014. [Online chapter 4].
Luscinia. Spectrogram window, Outubro 2017. [Online Parameters].
Geoffroy Peeters. A large set of audio features for sound description. Ircam - Analysis/Synthesis Team, 2004.
N. Fakotakis e G. Kokkinakis T. Ganchev. Comparative evaluation of various mfcc implementations on the speaker verification task. 10th International Conference on Speechand Computer (SPECOM 2005), 2005.
Brian McFee, Colin Raffel, Dawen Liang, Daniel Ellis,Matt McVicar, Eric Battenberg, and Oriol Nieto. librosa: Audio and music signal analysis in python. 14th Python in Science COnference, s/n(s/n), 2015.
Eli Bressert. SciPy and NumPy. O’Reilly Media, 2012.
Yotam Mann. The nsynth dataset, Abril 2017. [Online NSynth Dataset].
Jesse Engel, Cinjon Resnick, Adam Roberts, Sander Dieleman, Mohammad Norouzi, Douglas Eck, and Karen Simonyan. Neural audio synthesis of musical notes with wavenet autoencoders. In Proceedings of the 34th International Conference on Machine Learning-Volume 70, pages1068–1077. JMLR. org, 2017.
Raphael Campos.Machine Learning Beyond Deep Learning Árvores de decisão. [link]. Accessed: 2019-12-04.
Diego Lopez Yse. Towards data science: The complete guide to decision trees. Accessed: 2019-12-04.
Márcio Porto Basgalupp. Árvores de Decisão. PhD thesis, Universidade Estadual de Campinas - UNICAMP, 2010.
GIlbert Tanner. Towards data science: A guide to ensemble learning: Increase your accuracy by combining model outputs, 2019. Accessed: 2019-12-04.
G. E. P. BOX. NON-NORMALITY AND TESTS ONVARIANCES. Biometrika, 40(3-4):318–335, 12 1953.
Lara Haidar-Ahmad. Music and instrument classification using deep learning technics. In Proceedings of the Conference on Neural Information Processing Systems, pages 1–6. Nips.cc, 2017.
Janet Marques and Pedro Moreno. A study of musical instrument classification using gaussian mixture models andsupport vector machines. Cambridge Research Laboratory-Technical Report Series, 09 1999.
A. Eronen and A. Klapuri. Musical instrument recognition using cepstral coefficients and temporal features. In2000 IEEE International Conference on Acoustics, Speech,and Signal Processing. Proceedings (Cat. No.00CH37100),volume 2, pages II753–II756 vol.2, 2000.
Babak Toghiani-Rizi and Marcus Windmark. Musical instrument recognition using their distinctive characteristicsin artificial neural networks. ArXiv, abs/1705.04971, 2017.
Martin Vetterli Paolo Prandoni. Signal Processing for Communications (Communication and Information Sciences). EPFL Press, 2008.
Ivan Eiji Simurra. A utilização descritores de áudio à análise e composição musical assistidas por computador: um estudo de caso na obra labori ruinae, 2016.
Ivan Eiji Simurra. Análise musical assistida por descritores de Áudio: um estudo de caso da obra reflexões de jônatas manzolli, 2015.
Sound Analysis Pro Editorial. Wiener entropy, Outubro2014. [Online chapter 4].
Luscinia. Spectrogram window, Outubro 2017. [Online Parameters].
Geoffroy Peeters. A large set of audio features for sound description. Ircam - Analysis/Synthesis Team, 2004.
N. Fakotakis e G. Kokkinakis T. Ganchev. Comparative evaluation of various mfcc implementations on the speaker verification task. 10th International Conference on Speechand Computer (SPECOM 2005), 2005.
Brian McFee, Colin Raffel, Dawen Liang, Daniel Ellis,Matt McVicar, Eric Battenberg, and Oriol Nieto. librosa: Audio and music signal analysis in python. 14th Python in Science COnference, s/n(s/n), 2015.
Eli Bressert. SciPy and NumPy. O’Reilly Media, 2012.
Yotam Mann. The nsynth dataset, Abril 2017. [Online NSynth Dataset].
Jesse Engel, Cinjon Resnick, Adam Roberts, Sander Dieleman, Mohammad Norouzi, Douglas Eck, and Karen Simonyan. Neural audio synthesis of musical notes with wavenet autoencoders. In Proceedings of the 34th International Conference on Machine Learning-Volume 70, pages1068–1077. JMLR. org, 2017.
Raphael Campos.Machine Learning Beyond Deep Learning Árvores de decisão. [link]. Accessed: 2019-12-04.
Diego Lopez Yse. Towards data science: The complete guide to decision trees. Accessed: 2019-12-04.
Márcio Porto Basgalupp. Árvores de Decisão. PhD thesis, Universidade Estadual de Campinas - UNICAMP, 2010.
GIlbert Tanner. Towards data science: A guide to ensemble learning: Increase your accuracy by combining model outputs, 2019. Accessed: 2019-12-04.
G. E. P. BOX. NON-NORMALITY AND TESTS ONVARIANCES. Biometrika, 40(3-4):318–335, 12 1953.
Publicado
24/10/2021
Como Citar
VIEIRA, Rômulo; ARAÚJO, João Teixeira; BATISTA, Edimilson; SCHIAVONI, Flávio Luiz.
Automatic classification of instruments from supervised methods of machine learning. In: SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO MUSICAL (SBCM), 18. , 2021, Recife.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2021
.
p. 1-7.
DOI: https://doi.org/10.5753/sbcm.2021.19418.
