Title: Spectro-temporal features for environmental sound classification
Authors: Khine Zar Thwe; Mie Mie Thaw
Addresses: University of Computer Studies, Mandalay, Myanmar ' University of Computer Studies, Mandalay, Myanmar
Abstract: This paper proposes 2N_BILBP feature extraction method based on spectro-temporal features for sound event classification. Spectro-temporal features have a similar pattern to texture features in image processing. So, the concept of texture features is used in this digital signal processing field. Two-neighbour bidirectional local binary pattern (2N_BILBP) is used in this paper for feature extraction. 2N_BILBP is also compared with the previous method called bidirectional local binary pattern. Firstly, the input audio is converted into spectrogram using short-time Fourier transform and then gamma tone is used. The resulting gamma-tone-like spectrogram is then used to extract features. These features are used as feature features. Finally, the input audio is labelled using this feature vector. Evaluation is tested on three datasets called ESC-10 dataset, ESC-50 dataset and UrbanSound8K datasets called benchmark datasets.
Keywords: local binary pattern; sound event classification; audio event classification; texture features; spectro-temporal features; ESC-10 dataset; ESC-50 dataset; UrbanSound8K dataset.
DOI: 10.1504/IJCSE.2019.103816
International Journal of Computational Science and Engineering, 2019 Vol.20 No.2, pp.179 - 189
Received: 06 Apr 2018
Accepted: 18 Sep 2018
Published online: 29 Nov 2019 *