The Impact of Attention Mechanisms on Speech Emotion Recognition
Speech emotion recognition (SER) plays an important role in real-time applications of human-machine interaction. The Attention Mechanism is widely used to improve the performance of SER. However, the applicable rules of attention mechanism are not deeply discussed. This paper discussed the differenc...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/22/7530 |
| _version_ | 1827675453320593408 |
|---|---|
| author | Shouyan Chen Mingyan Zhang Xiaofen Yang Zhijia Zhao Tao Zou Xinqi Sun |
| author_facet | Shouyan Chen Mingyan Zhang Xiaofen Yang Zhijia Zhao Tao Zou Xinqi Sun |
| author_sort | Shouyan Chen |
| collection | DOAJ |
| description | Speech emotion recognition (SER) plays an important role in real-time applications of human-machine interaction. The Attention Mechanism is widely used to improve the performance of SER. However, the applicable rules of attention mechanism are not deeply discussed. This paper discussed the difference between Global-Attention and Self-Attention and explored their applicable rules to SER classification construction. The experimental results show that the Global-Attention can improve the accuracy of the sequential model, while the Self-Attention can improve the accuracy of the parallel model when conducting the model with the CNN and the LSTM. With this knowledge, a classifier (CNN-LSTM×2+Global-Attention model) for SER is proposed. The experiments result show that it could achieve an accuracy of 85.427% on the EMO-DB dataset. |
| first_indexed | 2024-03-10T05:05:12Z |
| format | Article |
| id | doaj.art-9c6f6ffb5d894251a31c7913baf9b515 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T05:05:12Z |
| publishDate | 2021-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-9c6f6ffb5d894251a31c7913baf9b5152023-11-23T01:24:55ZengMDPI AGSensors1424-82202021-11-012122753010.3390/s21227530The Impact of Attention Mechanisms on Speech Emotion RecognitionShouyan Chen0Mingyan Zhang1Xiaofen Yang2Zhijia Zhao3Tao Zou4Xinqi Sun5School of Mechanical and Electrical Engineering, Guangzhou University, Guangzhou 510006, ChinaSchool of Mechanical and Electrical Engineering, Guangzhou University, Guangzhou 510006, ChinaSchool of Mechanical and Electrical Engineering, Guangzhou University, Guangzhou 510006, ChinaSchool of Mechanical and Electrical Engineering, Guangzhou University, Guangzhou 510006, ChinaSchool of Mechanical and Electrical Engineering, Guangzhou University, Guangzhou 510006, ChinaSchool of Mechanical and Electrical Engineering, Guangzhou University, Guangzhou 510006, ChinaSpeech emotion recognition (SER) plays an important role in real-time applications of human-machine interaction. The Attention Mechanism is widely used to improve the performance of SER. However, the applicable rules of attention mechanism are not deeply discussed. This paper discussed the difference between Global-Attention and Self-Attention and explored their applicable rules to SER classification construction. The experimental results show that the Global-Attention can improve the accuracy of the sequential model, while the Self-Attention can improve the accuracy of the parallel model when conducting the model with the CNN and the LSTM. With this knowledge, a classifier (CNN-LSTM×2+Global-Attention model) for SER is proposed. The experiments result show that it could achieve an accuracy of 85.427% on the EMO-DB dataset.https://www.mdpi.com/1424-8220/21/22/7530artificial intelligencespeech emotion recognitionattention mechanismneural networks |
| spellingShingle | Shouyan Chen Mingyan Zhang Xiaofen Yang Zhijia Zhao Tao Zou Xinqi Sun The Impact of Attention Mechanisms on Speech Emotion Recognition Sensors artificial intelligence speech emotion recognition attention mechanism neural networks |
| title | The Impact of Attention Mechanisms on Speech Emotion Recognition |
| title_full | The Impact of Attention Mechanisms on Speech Emotion Recognition |
| title_fullStr | The Impact of Attention Mechanisms on Speech Emotion Recognition |
| title_full_unstemmed | The Impact of Attention Mechanisms on Speech Emotion Recognition |
| title_short | The Impact of Attention Mechanisms on Speech Emotion Recognition |
| title_sort | impact of attention mechanisms on speech emotion recognition |
| topic | artificial intelligence speech emotion recognition attention mechanism neural networks |
| url | https://www.mdpi.com/1424-8220/21/22/7530 |
| work_keys_str_mv | AT shouyanchen theimpactofattentionmechanismsonspeechemotionrecognition AT mingyanzhang theimpactofattentionmechanismsonspeechemotionrecognition AT xiaofenyang theimpactofattentionmechanismsonspeechemotionrecognition AT zhijiazhao theimpactofattentionmechanismsonspeechemotionrecognition AT taozou theimpactofattentionmechanismsonspeechemotionrecognition AT xinqisun theimpactofattentionmechanismsonspeechemotionrecognition AT shouyanchen impactofattentionmechanismsonspeechemotionrecognition AT mingyanzhang impactofattentionmechanismsonspeechemotionrecognition AT xiaofenyang impactofattentionmechanismsonspeechemotionrecognition AT zhijiazhao impactofattentionmechanismsonspeechemotionrecognition AT taozou impactofattentionmechanismsonspeechemotionrecognition AT xinqisun impactofattentionmechanismsonspeechemotionrecognition |