Rough Set-Based Classification of Audio Data
- 18 Downloads
For effective multimedia content, retrieval audio data plays an important role. Recognising classes of audio data which is neither music nor speech is a challenging task; in this aspect, the authors proposed to work on environment sounds. To represent the audio data, low-level features are extracted. These low-level descriptors are computed from both time domain and frequency domain representation of audio data. From the extracted descriptors, midterm statistics are computed and an information system (IS) is built with class labels. From this IS using the concept of rough set theory, reducts are computed, and from the reducts, rules are generated. The rules obtained are tested against the test set sampled from ESC-10 dataset.
KeywordsESC-10 Information system Discretisation Discernibility relation Reduct Rough set classifier
- 2.Piczak, K.J. 2015. ESC: Dataset for Environmental Sound Classification. In: 23rd Annual ACM Conference on Multimedia, Brisbane, Australia. pp. 1015–1018.Google Scholar
- 3.Florian, Eyben. (2016). Real-time Speech and Music Classification by Large Audio Feature Space Extraction. Springer Theses, Springer International Publishing.Google Scholar
- 4.Giannakopoulos, Theodoros, and Aggelos Pikrakis. 2014. Introduction to Audio Analysis: A MATLAB® Approach. Academic Press.Google Scholar
- 5.Piczak, K.J. https://github.com/karoldvl/ESC-50.
- 6.Sailor, Hardik B., Dharmesh M. Agrawal, and Hemant A. Patil. 2017. Unsupervised Filterbank Learning Using Convolutional Restricted Boltzmann Machine for Environmental Sound Classification. In: INTERSPEECH, 2017 Aug 2017, Stockholm, Sweden, pp. 3107–3111.Google Scholar
- 7.Tak, Rishabh N., Dharmesh M. Agrawal, and Hemant A. Patil. 2017. Novel Phase Encoded Mel Filterbank Energies for Environmental Sound Classification. In: Pattern Recognition and Machine Intelligence: 7th International Conference, PReMI 2017, Kolkata, India, 5–8 Dec 2017, pp. 317-325.Google Scholar
- 8.Tokozume, Yuji, and Tatsuya Harada. 2017. Learning Environmental Sounds with End-To-End Convolutional Neural Network. In 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, LA, pp. 2721–2725.Google Scholar
- 9.Piczak, K.J. 2015. Environmental Sound Classification with Convolutional Neural Networks. In 25th International Workshop on Machine Learning for Signal Processing (MLSP), pp. 1–6. Boston, MA, USA.Google Scholar
- 10.Tokozume, Yuji, Yoshitaka Ushiku, and Tatsuya Harada. 2018. Learning from Between-Class Examples for Deep Sound Recognition. In Sixth International Conference on Learning Representations, ICLR, Vancouver.Google Scholar
- 11.Baelde, Maxime, Christophe Biernacki, and Raphael Greff. 2017. A Mixture Model-Based Real-Time Audio Sources Classification Method. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, LA, 2017, pp. 2427–2431. (2017).Google Scholar
- 12.Freitag, Michael, et al. 2017. auDeep: Unsupervised Learning of Representations from Audio with Deep Recurrent Neural Networks. The Journal of Machine Learning Research 18 (1): 6340–6344.Google Scholar
- 13.Audio Files. Retrieved from https://freesound.org/.
- 14.Zdzisław, Pawlak. (1982). Rough Sets. International Journal of Computer and Information Sciences 11 (5).Google Scholar
- 15.Li, Xiao-Li, Zhen-Long Du, Tong Wang, and Dong-Mei Yu. Audio Feature Selection Based on Rough Set. International Journal of Information Technology 11 (6).Google Scholar