Experiment Science and Technology, Volume. 23, Issue 4, 1(2025)

Experimental Design of Speech Emotion Recognition with the Multi-Task Teacher-Student Model

Linhui SUN*, Ping’an LI, Yunlong LEI, and Zixiao ZHANG
Author Affiliations
  • School of Communications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210003, China
  • show less
    Figures & Tables(7)
    • Table 1. [in Chinese]

      View table
      View in Article

      Table 1. [in Chinese]

      卷积核输入通道输出通道步长
      TransCNN layer 1210245122
      TransCNN layer 225125122
      TransCNN layer 3310245122
      TransCNN layer 435125122
      TransCNN layer 5310245122
      TransCNN layer 635125122
      TransCNN layer 710102415
    • Table 2. [in Chinese]

      View table
      View in Article

      Table 2. [in Chinese]

      SNR/dB模型名称base(SER)教师模型多级连接语音增强WA/%UA/%
      10MCTM(ours)67.7268.98
      MCTM-TM××65.4565.19
      MCTM-TM-SEM×××64.0164.33
      MCTM-MECA×66.9567.72
      5MCTM(ours)65.0165.69
      MCTM-TM××63.2263.48
      MCTM-TM-SEM×××62.0662.34
      MCTM-MECA×64.4164.97
      0MCTM(ours)61.1061.26
      MCTM-TM××58.5358.16
      MCTM-TM-SEM×××57.9258.1
      MCTM-MECA×60.2460.40
      −5MCTM(ours)57.9558.10
      MCTM-TM××54.0554.41
      MCTM-TM-SEM×××53.7654.06
      MCTM-MECA×56.7756.93
      −10MCTM(ours)54.2354.73
      MCTM-TM×××52.6152.96
      MCTM-TM-SEM×××51.8852.21
      MCTM-MECA×53.7953.88
    • Table 3. [in Chinese]

      View table
      View in Article

      Table 3. [in Chinese]

      模型10 dB5 dB0 dB−5 dB−10 dB
      WAUAWAUAWAUAWAUAWAUA
      MHCNN[19]58.2656.9255.2752.3852.9748.2949.1246.946.0844.10
      AACNN[17]60.8559.3056.5153.4252.1849.2149.2947.0446.2644.76
      GLAM[20]59.9257.1556.9454.2354.5651.3750.3549.3046.9545.01
      GM-TCNet[21]61.1359.7158.6057.0556.3053.7251.7949.6048.7946.30
      TIM-Net[22]61.7460.4959.4357.2656.9254.6352.6150.9849.3147.09
      MCTM67.7268.9865.0165.6961.1061.2657.9558.1054.2354.73
    Tools

    Get Citation

    Copy Citation Text

    Linhui SUN, Ping’an LI, Yunlong LEI, Zixiao ZHANG. Experimental Design of Speech Emotion Recognition with the Multi-Task Teacher-Student Model[J]. Experiment Science and Technology, 2025, 23(4): 1

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Jul. 25, 2024

    Accepted: Oct. 30, 2024

    Published Online: Jul. 30, 2025

    The Author Email: Linhui SUN (sunlh@njupt.edu.cn)

    DOI:10.12179/1672-4550.20240391

    Topics