Abstract
This paper presents the usage of deep learning in flue pipe type recognition. The main thesis is the possibility of recognizing the type of labium based on the sound generated by the flue pipe. For the purpose of our work, we prepared a large data set of high-quality recordings, carried out in an organbuilder’s workshop. Very high accuracy has been achieved in our experiments on these data using Artificial Neural Networks (ANN), trained to recognize the details of the pipe mouth construction. The organbuilders claim that they can distinguish the pipe mouth type only by hearing it, and this is why we decided to verify if it is possible to train ANN to recognize the details of the organ pipe, as this confirms a possibility that a human sense of hearing may be trained as well. In the future, the usage of deep learning in the recognition of pipe sound parameters may be used in the voicing of the pipe organ and the selection of appropriate parameters of pipes to obtain the desired timbre.
Keywords
- Flue pipe
- Deep learning
- Labium recognition
Partially supported by research funds sponsored by the Ministry of Science and Higher Education in Poland.
This is a preview of subscription content, access via your institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Angster, J., Rusz, P., Miklos, A.: Acoustics of organ pipes and future trends in the research. Acoust. Today 1(13), 10–18 (2017)
Außerlechner, H., Trommer, T., Angster, J., Miklos, A.: Experimental jet velocity and edge tone investigations on a foot model of an organ pipe. J. Acoust. Soc. Am. 2(126), 878–886 (2009). https://doi.org/10.1121/1.3158935
Deep Cognition Homepage. https://deepcognition.ai. Accessed 24 Apr 2020
Herremans, D., Chuan, C.: The emergence of deep learning: new opportunities for music and audio technologies. Neural Comput. Appl. 32, 913–914 (2020). https://doi.org/10.1007/s00521-019-04166-0
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997). https://doi.org/10.1162/neco.1997.9.8.1735
Hruška, V., Dlask, P.: Investigation of the sound source regions in open and closed organ pipes. Arch. Acoust. 3(44), 467–474 (2019). https://doi.org/10.24425/aoa.2019.129262
Hruška, V., Dlask, P.: Connections between organ pipe noise and Shannon entropy of the airflow: preliminary results. Acta Acustica United Acustica 103, 1100–1105 (2017)
Humphrey, E.J., Bello, J.P., LeCun, Y.: Feature learning and deep architectures: new directions for music informatics. J. Intell. Inf. Syst. 41, 461–481 (2013). https://doi.org/10.1007/s10844-013-0248-5
Koutini, K., Chowdhury, S., Haunschmid, V., Eghbal-zadeh H., Widmer, G.: Emotion and theme recognition in music with frequency-aware RF-regularized CNNs. MediaEval 1919, 27–29 October 2019. ArXiv abs/1911.05833. Sophia Antipolis (2019)
Krizhevsky, A., Sutskever, I., Hinton, G.: ImageNet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012). https://doi.org/10.1145/3065386
Lathi, B.P.: Linear systems and signals, 2nd edn. Oxford University Press, New York (2010)
Lehner, B., Widmer, G., Bock., S.: A low-latency, real-time-capable singing voice detection method with LSTM recurrent neural networks. In: 2015 23\(^{rd}\) European Signal Processing Conference (EUSIPCO), Nice, pp. 21 – 25 (2015). https://doi.org/10.1109/EUSIPCO.2015.7362337
Rucz, P., Augusztinovicz, F., Angster, J., Preukschat, T., Miklos, A.: Acoustic behaviour of tuning slots of labial organ pipes.J. Acoust. Soc. Am. 5(135), 3056–3065 (2014). https://doi.org/10.1121/1.4869679
Sakamoto, Y., Yoshikawa, S., Angster, J.: Acoustical investigations on the ears of flue or-GAN pipes. In: Forum Acusticum, pp. 647-651. EAA-Opakfi Hungary, Budapest (2005)
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2014). https://doi.org/10.5555/2627435.2670313
Vaik, I., Paal, G.: Flow simulations on an organ pipe foot model. J. Acoust. Soc. Am. 2(133), 1102–1110 (2013). https://doi.org/10.1121/1.4773861
Verge, M., Fabre, B., Mahu, W., Hirschberg, A., et al.: Jet formation and jet velocity fluctuations in a flue organ pipe. J. Acoust. Soc. Am. 2(95), 1119–1132 (1994). https://doi.org/10.1121/1.408460
Widmer, G.: On the potential of machine learning for music research. In: Miranda, E.R. (ed.) Readings in Music and Artificial Intelligence. Routledge, New York (2013)
Zeiler, M.: ADADELTA: an adaptive learning rate method. https://arxiv.org/abs/1212.5701. Accessed 24 Apr 2020
Acknowledgements
Special thanks to the organbuilder Władysław Cepka for his invaluable help and providing the workshop and organ pipes for sound recording.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Węgrzyn, D., Wrzeciono, P., Wieczorkowska, A. (2021). Recognition of the Flue Pipe Type Using Deep Learning. In: Stettinger, M., Leitner, G., Felfernig, A., Ras, Z.W. (eds) Intelligent Systems in Industrial Applications. ISMIS 2020. Studies in Computational Intelligence, vol 949. Springer, Cham. https://doi.org/10.1007/978-3-030-67148-8_7
Download citation
DOI: https://doi.org/10.1007/978-3-030-67148-8_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-67147-1
Online ISBN: 978-3-030-67148-8
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)