Efficient Noisy Sound-Event Mixture Classification Using Adaptive-Sparse Complex-Valued Matrix Factorization and OvsO SVM
This paper proposes a solution for events classification from a sole noisy mixture that consist of two major steps: a sound-event separation and a sound-event classification. The traditional complex nonnegative matrix factorization (CMF) is extended by cooperation with the optimal adaptive <i>...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/16/4368 |
| _version_ | 1797560254090706944 |
|---|---|
| author | Phetcharat Parathai Naruephorn Tengtrairat Wai Lok Woo Mohammed A. M. Abdullah Gholamreza Rafiee Ossama Alshabrawy |
| author_facet | Phetcharat Parathai Naruephorn Tengtrairat Wai Lok Woo Mohammed A. M. Abdullah Gholamreza Rafiee Ossama Alshabrawy |
| author_sort | Phetcharat Parathai |
| collection | DOAJ |
| description | This paper proposes a solution for events classification from a sole noisy mixture that consist of two major steps: a sound-event separation and a sound-event classification. The traditional complex nonnegative matrix factorization (CMF) is extended by cooperation with the optimal adaptive <i>L</i><sub>1</sub> sparsity to decompose a noisy single-channel mixture. The proposed adaptive <i>L</i><sub>1</sub> sparsity CMF algorithm encodes the spectra pattern and estimates the phase of the original signals in time-frequency representation. Their features enhance the temporal decomposition process efficiently. The support vector machine (SVM) based one versus one (OvsO) strategy was applied with a mean supervector to categorize the demixed sound into the matching sound-event class. The first step of the multi-class MSVM method is to segment the separated signal into blocks by sliding demixed signals, then encoding the three features of each block. Mel frequency cepstral coefficients, short-time energy, and short-time zero-crossing rate are learned with multi sound-event classes by the SVM based OvsO method. The mean supervector is encoded from the obtained features. The proposed method has been evaluated with both separation and classification scenarios using real-world single recorded signals and compared with the state-of-the-art separation method. Experimental results confirmed that the proposed method outperformed the state-of-the-art methods. |
| first_indexed | 2024-03-10T17:56:51Z |
| format | Article |
| id | doaj.art-45fa3ea579e8460a838abb0e6865b9b8 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T17:56:51Z |
| publishDate | 2020-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-45fa3ea579e8460a838abb0e6865b9b82023-11-20T09:09:23ZengMDPI AGSensors1424-82202020-08-012016436810.3390/s20164368Efficient Noisy Sound-Event Mixture Classification Using Adaptive-Sparse Complex-Valued Matrix Factorization and OvsO SVMPhetcharat Parathai0Naruephorn Tengtrairat1Wai Lok Woo2Mohammed A. M. Abdullah3Gholamreza Rafiee4Ossama Alshabrawy5School of Software Engineering, Payap University, Chiang Mai 50000, ThailandSchool of Software Engineering, Payap University, Chiang Mai 50000, ThailandDepartment of Computer and Information Sciences, Northumbria University, Newcastle upon Tyne NE1 8ST, UKComputer and Information Engineering Department, Ninevah University, Mosul 41002, IraqSchool of Electronics, Electrical Engineering and Computer Science, Queen’s University Belfast, Belfast BT9 5BN, UKDepartment of Computer and Information Sciences, Northumbria University, Newcastle upon Tyne NE1 8ST, UKThis paper proposes a solution for events classification from a sole noisy mixture that consist of two major steps: a sound-event separation and a sound-event classification. The traditional complex nonnegative matrix factorization (CMF) is extended by cooperation with the optimal adaptive <i>L</i><sub>1</sub> sparsity to decompose a noisy single-channel mixture. The proposed adaptive <i>L</i><sub>1</sub> sparsity CMF algorithm encodes the spectra pattern and estimates the phase of the original signals in time-frequency representation. Their features enhance the temporal decomposition process efficiently. The support vector machine (SVM) based one versus one (OvsO) strategy was applied with a mean supervector to categorize the demixed sound into the matching sound-event class. The first step of the multi-class MSVM method is to segment the separated signal into blocks by sliding demixed signals, then encoding the three features of each block. Mel frequency cepstral coefficients, short-time energy, and short-time zero-crossing rate are learned with multi sound-event classes by the SVM based OvsO method. The mean supervector is encoded from the obtained features. The proposed method has been evaluated with both separation and classification scenarios using real-world single recorded signals and compared with the state-of-the-art separation method. Experimental results confirmed that the proposed method outperformed the state-of-the-art methods.https://www.mdpi.com/1424-8220/20/16/4368audio signal processingsound event classificationnonnegative matric factorizationblind signal separationsupport vector machines |
| spellingShingle | Phetcharat Parathai Naruephorn Tengtrairat Wai Lok Woo Mohammed A. M. Abdullah Gholamreza Rafiee Ossama Alshabrawy Efficient Noisy Sound-Event Mixture Classification Using Adaptive-Sparse Complex-Valued Matrix Factorization and OvsO SVM Sensors audio signal processing sound event classification nonnegative matric factorization blind signal separation support vector machines |
| title | Efficient Noisy Sound-Event Mixture Classification Using Adaptive-Sparse Complex-Valued Matrix Factorization and OvsO SVM |
| title_full | Efficient Noisy Sound-Event Mixture Classification Using Adaptive-Sparse Complex-Valued Matrix Factorization and OvsO SVM |
| title_fullStr | Efficient Noisy Sound-Event Mixture Classification Using Adaptive-Sparse Complex-Valued Matrix Factorization and OvsO SVM |
| title_full_unstemmed | Efficient Noisy Sound-Event Mixture Classification Using Adaptive-Sparse Complex-Valued Matrix Factorization and OvsO SVM |
| title_short | Efficient Noisy Sound-Event Mixture Classification Using Adaptive-Sparse Complex-Valued Matrix Factorization and OvsO SVM |
| title_sort | efficient noisy sound event mixture classification using adaptive sparse complex valued matrix factorization and ovso svm |
| topic | audio signal processing sound event classification nonnegative matric factorization blind signal separation support vector machines |
| url | https://www.mdpi.com/1424-8220/20/16/4368 |
| work_keys_str_mv | AT phetcharatparathai efficientnoisysoundeventmixtureclassificationusingadaptivesparsecomplexvaluedmatrixfactorizationandovsosvm AT naruephorntengtrairat efficientnoisysoundeventmixtureclassificationusingadaptivesparsecomplexvaluedmatrixfactorizationandovsosvm AT wailokwoo efficientnoisysoundeventmixtureclassificationusingadaptivesparsecomplexvaluedmatrixfactorizationandovsosvm AT mohammedamabdullah efficientnoisysoundeventmixtureclassificationusingadaptivesparsecomplexvaluedmatrixfactorizationandovsosvm AT gholamrezarafiee efficientnoisysoundeventmixtureclassificationusingadaptivesparsecomplexvaluedmatrixfactorizationandovsosvm AT ossamaalshabrawy efficientnoisysoundeventmixtureclassificationusingadaptivesparsecomplexvaluedmatrixfactorizationandovsosvm |