Binaural Acoustic Scene Classification Using Wavelet Scattering, Parallel Ensemble Classifiers and Nonlinear Fusion
The analysis of ambient sounds can be very useful when developing sound base intelligent systems. Acoustic scene classification (ASC) is defined as identifying the area of a recorded sound or clip among some predefined scenes. ASC has huge potential to be used in urban sound event classification sys...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/4/1535 |
| _version_ | 1797476719164129280 |
|---|---|
| author | Vahid Hajihashemi Abdorreza Alavi Gharahbagh Pedro Miguel Cruz Marta Campos Ferreira José J. M. Machado João Manuel R. S. Tavares |
| author_facet | Vahid Hajihashemi Abdorreza Alavi Gharahbagh Pedro Miguel Cruz Marta Campos Ferreira José J. M. Machado João Manuel R. S. Tavares |
| author_sort | Vahid Hajihashemi |
| collection | DOAJ |
| description | The analysis of ambient sounds can be very useful when developing sound base intelligent systems. Acoustic scene classification (ASC) is defined as identifying the area of a recorded sound or clip among some predefined scenes. ASC has huge potential to be used in urban sound event classification systems. This research presents a hybrid method that includes a novel mathematical fusion step which aims to tackle the challenges of ASC accuracy and adaptability of current state-of-the-art models. The proposed method uses a stereo signal, two ensemble classifiers (random subspace), and a novel mathematical fusion step. In the proposed method, a stable, invariant signal representation of the stereo signal is built using Wavelet Scattering Transform (WST). For each mono, i.e., left and right, channel, a different random subspace classifier is trained using WST. A novel mathematical formula for fusion step was developed, its parameters being found using a Genetic algorithm. The results on the DCASE 2017 dataset showed that the proposed method has higher classification accuracy (about 95%), pushing the boundaries of existing methods. |
| first_indexed | 2024-03-09T21:06:38Z |
| format | Article |
| id | doaj.art-e6cd1177e9344346bbb73667fb337b26 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T21:06:38Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-e6cd1177e9344346bbb73667fb337b262023-11-23T22:01:06ZengMDPI AGSensors1424-82202022-02-01224153510.3390/s22041535Binaural Acoustic Scene Classification Using Wavelet Scattering, Parallel Ensemble Classifiers and Nonlinear FusionVahid Hajihashemi0Abdorreza Alavi Gharahbagh1Pedro Miguel Cruz2Marta Campos Ferreira3José J. M. Machado4João Manuel R. S. Tavares5Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, s/n, 4200-465 Porto, PortugalFaculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, s/n, 4200-465 Porto, PortugalBosch Security Systems S.A., EN109-Zona Industrial de Ovar, 3880-080 Ovar, PortugalFaculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, s/n, 4200-465 Porto, PortugalDepartamento de Engenharia Mecânica, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, s/n, 4200-465 Porto, PortugalDepartamento de Engenharia Mecânica, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, s/n, 4200-465 Porto, PortugalThe analysis of ambient sounds can be very useful when developing sound base intelligent systems. Acoustic scene classification (ASC) is defined as identifying the area of a recorded sound or clip among some predefined scenes. ASC has huge potential to be used in urban sound event classification systems. This research presents a hybrid method that includes a novel mathematical fusion step which aims to tackle the challenges of ASC accuracy and adaptability of current state-of-the-art models. The proposed method uses a stereo signal, two ensemble classifiers (random subspace), and a novel mathematical fusion step. In the proposed method, a stable, invariant signal representation of the stereo signal is built using Wavelet Scattering Transform (WST). For each mono, i.e., left and right, channel, a different random subspace classifier is trained using WST. A novel mathematical formula for fusion step was developed, its parameters being found using a Genetic algorithm. The results on the DCASE 2017 dataset showed that the proposed method has higher classification accuracy (about 95%), pushing the boundaries of existing methods.https://www.mdpi.com/1424-8220/22/4/1535urban sounds classificationstereo signalsound base intelligent systemmachine learninggenetic algorithm |
| spellingShingle | Vahid Hajihashemi Abdorreza Alavi Gharahbagh Pedro Miguel Cruz Marta Campos Ferreira José J. M. Machado João Manuel R. S. Tavares Binaural Acoustic Scene Classification Using Wavelet Scattering, Parallel Ensemble Classifiers and Nonlinear Fusion Sensors urban sounds classification stereo signal sound base intelligent system machine learning genetic algorithm |
| title | Binaural Acoustic Scene Classification Using Wavelet Scattering, Parallel Ensemble Classifiers and Nonlinear Fusion |
| title_full | Binaural Acoustic Scene Classification Using Wavelet Scattering, Parallel Ensemble Classifiers and Nonlinear Fusion |
| title_fullStr | Binaural Acoustic Scene Classification Using Wavelet Scattering, Parallel Ensemble Classifiers and Nonlinear Fusion |
| title_full_unstemmed | Binaural Acoustic Scene Classification Using Wavelet Scattering, Parallel Ensemble Classifiers and Nonlinear Fusion |
| title_short | Binaural Acoustic Scene Classification Using Wavelet Scattering, Parallel Ensemble Classifiers and Nonlinear Fusion |
| title_sort | binaural acoustic scene classification using wavelet scattering parallel ensemble classifiers and nonlinear fusion |
| topic | urban sounds classification stereo signal sound base intelligent system machine learning genetic algorithm |
| url | https://www.mdpi.com/1424-8220/22/4/1535 |
| work_keys_str_mv | AT vahidhajihashemi binauralacousticsceneclassificationusingwaveletscatteringparallelensembleclassifiersandnonlinearfusion AT abdorrezaalavigharahbagh binauralacousticsceneclassificationusingwaveletscatteringparallelensembleclassifiersandnonlinearfusion AT pedromiguelcruz binauralacousticsceneclassificationusingwaveletscatteringparallelensembleclassifiersandnonlinearfusion AT martacamposferreira binauralacousticsceneclassificationusingwaveletscatteringparallelensembleclassifiersandnonlinearfusion AT josejmmachado binauralacousticsceneclassificationusingwaveletscatteringparallelensembleclassifiersandnonlinearfusion AT joaomanuelrstavares binauralacousticsceneclassificationusingwaveletscatteringparallelensembleclassifiersandnonlinearfusion |