Scholarlink

Coal mining is one of the industries with a relatively high frequency of work-related accidents. By using speech emotion recognition technology to monitor miners’ unsafe emotions in real-time, appropriate measures can be taken promptly to eliminate potential safety hazards. This paper proposes a two-stage spontaneous speech emotion recognition method. Through transfer learning technology, it utilizes the intermediate values of multi-emotion intensity models to optimize the recognition effect of the dominant emotion model, thereby more effectively capturing comprehensive emotional information and improving the recognition accuracy of the dominant emotion.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Two-Stage Speech Emotion Recognition Method for Coal Mining Applications

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Xiaohua Zhao,
  • Lihua Xu,
  • Ronghua Zhang,
  • Min Gong,
  • Guoyuan Lin,
  • Wei Chen,
  • Hengbo Li,
  • Yingchun Liu

摘要

Coal mining is one of the industries with a relatively high frequency of work-related accidents. By using speech emotion recognition technology to monitor miners’ unsafe emotions in real-time, appropriate measures can be taken promptly to eliminate potential safety hazards. This paper proposes a two-stage spontaneous speech emotion recognition method. Through transfer learning technology, it utilizes the intermediate values of multi-emotion intensity models to optimize the recognition effect of the dominant emotion model, thereby more effectively capturing comprehensive emotional information and improving the recognition accuracy of the dominant emotion.