About | Contact Us | Register | Login
ProceedingsSeriesJournalsSearchEAI
Mobile Multimedia Communications. 16th EAI International Conference, MobiMedia 2023, Guilin, China, July 22-24, 2023, Proceedings

Research Article

Frame Optimization in Speech Emotion Recognition Based on Improved EMD and SVM Algorithms

Cite
BibTeX Plain Text
  • @INPROCEEDINGS{10.1007/978-3-031-60347-1_11,
        author={Chuan-Jie Guo and Shu-Ya Jin and Yu-Zhe Zhang and Chi-Yuan Ma and Muhammad Adeel and Zhi-Yong Tao},
        title={Frame Optimization in Speech Emotion Recognition Based on Improved EMD and SVM Algorithms},
        proceedings={Mobile Multimedia Communications. 16th EAI International Conference, MobiMedia 2023, Guilin, China, July 22-24, 2023, Proceedings},
        proceedings_a={MOBIMEDIA},
        year={2024},
        month={10},
        keywords={Speech emotion recognition Improved EMD Signal framing SVM},
        doi={10.1007/978-3-031-60347-1_11}
    }
    
  • Chuan-Jie Guo
    Shu-Ya Jin
    Yu-Zhe Zhang
    Chi-Yuan Ma
    Muhammad Adeel
    Zhi-Yong Tao
    Year: 2024
    Frame Optimization in Speech Emotion Recognition Based on Improved EMD and SVM Algorithms
    MOBIMEDIA
    Springer
    DOI: 10.1007/978-3-031-60347-1_11
Chuan-Jie Guo1, Shu-Ya Jin1, Yu-Zhe Zhang1, Chi-Yuan Ma1, Muhammad Adeel1, Zhi-Yong Tao1,*
  • 1: Guangxi Key Laboratory of Cognitive Radio and Information Processing, Ministry of Education, Guilin University of Electronic Technology
*Contact email: zytao@guet.edu.cn

Abstract

Emotional features of speech signals are one of the keys to human-computer interaction. However, there are still great difficulties and chances to extract emotional features. There is also great controversy regarding the part of signal preprocessing. This study divides the speech signal into small frames that overlap with a portion of the previous frame and adopts an improved empirical mode decomposition (EMD) based feature extraction method. The aim is to find the most suitable framing method. Each frame signal is processed by an improved EMD to generate a set of intrinsic mode functions (IMFs). Multidimensional features are extracted by calculating the central frequency and energy intensity of each IMF, and subsequently processing the center frequency of each IMF. Specifically, we focus on the top three IMFs in terms of energy intensity. Based on the improved algorithm, we investigate the effects of different frame lengths and frame shifts on the recognition rates of three emotion classifications: happy, angry, and sad. We find that the proposed method can reach the highest recognition rate when we use a 30 ms frame length with a 25% frame shift to separate the signals.

Keywords
Speech emotion recognition Improved EMD Signal framing SVM
Published
2024-10-25
Appears in
SpringerLink
http://dx.doi.org/10.1007/978-3-031-60347-1_11
Copyright © 2023–2025 ICST
EBSCOProQuestDBLPDOAJPortico
EAI Logo

About EAI

  • Who We Are
  • Leadership
  • Research Areas
  • Partners
  • Media Center

Community

  • Membership
  • Conference
  • Recognition
  • Sponsor Us

Publish with EAI

  • Publishing
  • Journals
  • Proceedings
  • Books
  • EUDL