Genetic Algorithm for High-Dimensional Emotion Recognition from Speech Signals
Feature selection plays a crucial role in establishing an effective speech emotion recognition system. To improve recognition accuracy, people always extract as many features as possible from speech signals. However, this may reduce efficiency. We propose a hybrid filter–wrapper feature selection ba...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/23/4779 |
| _version_ | 1797400246513303552 |
|---|---|
| author | Liya Yue Pei Hu Shu-Chuan Chu Jeng-Shyang Pan |
| author_facet | Liya Yue Pei Hu Shu-Chuan Chu Jeng-Shyang Pan |
| author_sort | Liya Yue |
| collection | DOAJ |
| description | Feature selection plays a crucial role in establishing an effective speech emotion recognition system. To improve recognition accuracy, people always extract as many features as possible from speech signals. However, this may reduce efficiency. We propose a hybrid filter–wrapper feature selection based on a genetic algorithm specifically designed for high-dimensional (HGA) speech emotion recognition. The algorithm first utilizes Fisher Score and information gain to comprehensively rank acoustic features, and then these features are assigned probabilities for inclusion in subsequent operations according to their ranking. HGA improves population diversity and local search ability by modifying the initial population generation method of genetic algorithm (GA) and introducing adaptive crossover and a new mutation strategy. The proposed algorithm clearly reduces the number of selected features in four common English speech emotion datasets. It is confirmed by K-nearest neighbor and random forest classifiers that it is superior to state-of-the-art algorithms in accuracy, precision, recall, and F1-Score. |
| first_indexed | 2024-03-09T01:52:47Z |
| format | Article |
| id | doaj.art-4bdb4e68e06d447590027baf0f98731f |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-09T01:52:47Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-4bdb4e68e06d447590027baf0f98731f2023-12-08T15:14:00ZengMDPI AGElectronics2079-92922023-11-011223477910.3390/electronics12234779Genetic Algorithm for High-Dimensional Emotion Recognition from Speech SignalsLiya Yue0Pei Hu1Shu-Chuan Chu2Jeng-Shyang Pan3Fanli Business School, Nanyang Institute of Technology, Nanyang 473004, ChinaSchool of Computer and Software, Nanyang Institute of Technology, Nanyang 473004, ChinaCollege of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaCollege of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaFeature selection plays a crucial role in establishing an effective speech emotion recognition system. To improve recognition accuracy, people always extract as many features as possible from speech signals. However, this may reduce efficiency. We propose a hybrid filter–wrapper feature selection based on a genetic algorithm specifically designed for high-dimensional (HGA) speech emotion recognition. The algorithm first utilizes Fisher Score and information gain to comprehensively rank acoustic features, and then these features are assigned probabilities for inclusion in subsequent operations according to their ranking. HGA improves population diversity and local search ability by modifying the initial population generation method of genetic algorithm (GA) and introducing adaptive crossover and a new mutation strategy. The proposed algorithm clearly reduces the number of selected features in four common English speech emotion datasets. It is confirmed by K-nearest neighbor and random forest classifiers that it is superior to state-of-the-art algorithms in accuracy, precision, recall, and F1-Score.https://www.mdpi.com/2079-9292/12/23/4779feature selectionspeech emotion recognitiongenetic algorithmhigh-dimensional |
| spellingShingle | Liya Yue Pei Hu Shu-Chuan Chu Jeng-Shyang Pan Genetic Algorithm for High-Dimensional Emotion Recognition from Speech Signals Electronics feature selection speech emotion recognition genetic algorithm high-dimensional |
| title | Genetic Algorithm for High-Dimensional Emotion Recognition from Speech Signals |
| title_full | Genetic Algorithm for High-Dimensional Emotion Recognition from Speech Signals |
| title_fullStr | Genetic Algorithm for High-Dimensional Emotion Recognition from Speech Signals |
| title_full_unstemmed | Genetic Algorithm for High-Dimensional Emotion Recognition from Speech Signals |
| title_short | Genetic Algorithm for High-Dimensional Emotion Recognition from Speech Signals |
| title_sort | genetic algorithm for high dimensional emotion recognition from speech signals |
| topic | feature selection speech emotion recognition genetic algorithm high-dimensional |
| url | https://www.mdpi.com/2079-9292/12/23/4779 |
| work_keys_str_mv | AT liyayue geneticalgorithmforhighdimensionalemotionrecognitionfromspeechsignals AT peihu geneticalgorithmforhighdimensionalemotionrecognitionfromspeechsignals AT shuchuanchu geneticalgorithmforhighdimensionalemotionrecognitionfromspeechsignals AT jengshyangpan geneticalgorithmforhighdimensionalemotionrecognitionfromspeechsignals |